HK1207776B - Method and apparatus for coded picture buffer removal times - Google Patents

Description

Method and device for coded picture buffer removal time

This application claims priority to U.S. Provisional Application No. 61/705,119, filed September 24, 2012, and U.S. Provisional Application No. 61/708,475, filed October 1, 2012, each of which is incorporated by reference in its entirety.

Technical Field

The present invention relates to video decoding.

Background Art

Digital video capabilities can be incorporated into a wide range of devices, including digital televisions, digital live broadcast systems, wireless broadcast systems, personal digital assistants (PDAs), laptop or desktop computers, tablet computers, e-book readers, digital cameras, digital recording devices, digital media players, video game devices, video game consoles, cellular or satellite radio telephones, so-called "smartphones," video teleconferencing devices, video streaming devices, and the like. Digital video devices implement video compression techniques, such as those described in the standards described by MPEG-2, MPEG-4, ITU-T H.263, ITU-T H.264/MPEG-4 Part 10 Advanced Video Coding (AVC), the High Efficiency Video Coding (HEVC) standard, and extensions of such standards. By implementing such video compression techniques, video devices can more efficiently transmit, receive, encode, decode, and/or store digital video information.

Video compression techniques perform spatial (intra-picture) prediction and/or temporal (inter-picture) prediction to reduce or remove redundancy inherent in video sequences. For block-based video coding, a video slice (i.e., a video frame or portion of a video frame) may be partitioned into video blocks, which may also be referred to as treeblocks, coding units (CUs), and/or coding nodes. Video blocks in an intra-coded (I) slice of a picture are encoded using spatial prediction with respect to reference samples in neighboring blocks in the same picture. Video blocks in an inter-coded (P or B) slice of a picture may use spatial prediction with respect to reference samples in neighboring blocks in the same picture or temporal prediction with respect to reference samples in other reference pictures. Pictures may be referred to as frames, and reference pictures may be referred to as reference frames.

Spatial or temporal prediction results in a predictive block for the block to be coded. Residual data represents the pixel differences between the original block to be coded and the predictive block. Inter-coded blocks are encoded according to motion vectors pointing to a block of reference samples forming the predictive block, and the residual data indicates the difference between the coded block and the predictive block. Intra-coded blocks are encoded according to an intra-coding mode and the residual data. For further compression, the residual data can be transformed from the pixel domain to the transform domain, resulting in residual transform coefficients that can then be quantized. The quantized transform coefficients, initially arranged in a two-dimensional array, can be scanned to produce a one-dimensional vector of transform coefficients, and entropy coding can be applied to achieve even more compression.

Summary of the Invention

In general, the techniques described in this disclosure relate to signaling and derivation of coded picture buffer removal times in video coding.

In one example, the techniques described in this disclosure relate to a method for decoding video data. The method may include decoding a duration between a coded picture buffer (CPB) removal time of a first decoding unit (DU) and a CPB removal time of a second DU in an access unit (AU), wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU. The method may further include determining a removal time of the first DU based at least in part on the decoded duration and decoding video data of the first DU based at least in part on the removal time.

In another example, the techniques described in this disclosure relate to a method for encoding video data. The method may include encoding a duration between a CPB removal time of a first DU and a CPB removal time of a second DU in an AU, wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU. The method may further include determining a removal time for the first DU based at least in part on the encoded duration.

In another example, a video coding device is provided that includes a video coder configured to code a duration between CPB removal times of a first DU and a second DU in an AU, wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU. The video coder is further configured to determine a removal time of the DU based at least on the coded duration.

The techniques described herein also include an example of a computer-readable storage medium having stored thereon instructions that, when executed, cause a processor of a device for decoding video data to code a duration between CPB removal times of a first DU and a second DU in an AU, wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU. The instructions, when executed, further cause the processor to determine a removal time for the DU based at least on the coded duration.

In another example, the techniques described in this disclosure relate to a video coding device. The video coding device may include means for coding a duration between a first decoding unit (DU) and a coded picture buffer (CPB) removal time of a second DU in an access unit (AU), wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU. The video coding device may further include means for determining a removal time of the DU based at least on the coded duration.

Such example techniques may be implemented together or separately.The techniques of this disclosure are also described in terms of an apparatus configured to implement the techniques and a computer-readable storage medium storing instructions that cause one or more processors to perform the techniques.

The details of one or more examples are set forth in the accompanying drawings and the description below. Other features, objects, and advantages will be apparent from the description and drawings, and from the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

1 is a block diagram illustrating an example video encoding and decoding system that may utilize the techniques described in this disclosure.

2 is a block diagram illustrating an example video encoder that may implement the techniques described in this disclosure.

3 is a block diagram illustrating an example video decoder that may implement the techniques described in this disclosure.

4 is a conceptual diagram illustrating two access units (AUs) in consecutive decoding order, which may have decoding times determined according to the techniques described in this disclosure.

5 is a flow diagram illustrating a method for determining a coded picture buffer (CPB) removal time for a first decoding unit (DU) in an AU based on the CPB removal time of a second DU of the AU, according to the techniques described in this disclosure.

6 is a flowchart illustrating another method for determining a coded picture buffer (CPB) removal time for a first decoding unit in an access unit based on the CPB removal time of a second decoding unit of the access unit, according to the techniques described in this disclosure.

7 is a flow diagram illustrating a method for deriving a CPB removal time for a first DU based at least in part on a sub-picture timing SEI message, according to the techniques described in this disclosure.

8 is a flow diagram illustrating another method for deriving a CPB removal time for a first DU based at least in part on an encoded sub-picture timing SEI message, according to the techniques described in this disclosure.

9 is a flowchart illustrating a method for decoding sequence level flags for sub-picture level coded picture buffer parameters according to the techniques described in this disclosure.

10 is a flowchart illustrating a method for encoding sequence level flags for sub-picture level coded picture buffer parameters according to the techniques described in this disclosure.

11 is a flowchart illustrating a method for decoding a DU with an extended definition according to the techniques described in this disclosure.

12 is a flowchart illustrating a method for encoding a DU with an extended definition according to the techniques described in this disclosure.

13 is a flow diagram illustrating a method for decoding buffering periods according to the techniques described in this disclosure.

14 is a flow diagram illustrating a method for encoding a buffering period according to the techniques described in this disclosure.

15 is a flowchart illustrating a method for decoding coded picture buffer arrival and nominal removal times according to the techniques described in this disclosure.

16 is a flowchart illustrating a method for encoding coded picture buffer arrival and nominal removal times according to the techniques described in this disclosure.

DETAILED DESCRIPTION

This disclosure describes techniques for error-resilient and efficient signaling and derivation of coded picture buffer (CPB) removal times for coded data units in video coding. The CPB removal time is also referred to as the decoding time. This disclosure provides techniques for determining the CPB removal time for a decoding unit (DU) of an access unit (AU) independent of the removal times of any other AUs. For example, the CPB removal time for a current DU of an AU is signaled based on the duration between the CPB removal time of the next DU in decoding order in the AU and the current DU, or the duration between the CPB removal time of the last DU in the AU and the current DU. In another example, CPB removal time derivation is specified using information carried in a sub-picture timing supplemental enhancement information (SEI) message. The duration between the CPB removal time of the last DU in the AU in decoding order and the DU associated with the sub-picture timing SEI message is signaled.

In addition, according to the techniques described herein, techniques are provided that include a sequence-level flag that can be signaled to indicate that sub-picture CPB parameters are present in only one of a picture timing SEI message or a sub-picture timing SEI message, but never in both. A flag equal to 1 indicates that the sub-picture-level CPB removal delay parameters are present in the picture timing SEI message and that no sub-picture timing SEI message is present. A flag equal to 0 indicates that the sub-picture-level CPB removal delay parameters are present in the sub-picture timing SEI message and that the picture timing SEI message does not include the sub-picture-level CPB removal delay parameters.

This disclosure also provides techniques for extending the definition of decoding units. It further provides techniques for restricting buffering period SEI messages and recovery point SEI messages so that they cannot be associated with AUs with a variable TemporalId greater than 0. The variable TemporalId is derived from syntax elements associated with each AU. It also provides techniques for providing a flag to signal whether CPB removal time is derived at the AU level or the sub-picture level.

The techniques described herein may be applied to various video coding standards, including ITU-T H.261, ISO/IEC MPEG-1 Visual, ITU-T H.262 or ISO/IEC MPEG-2 Visual, ITU-T H.263, ISO/IEC MPEG-4 Visual, and ITU-T H.264 (also known as ISO/IEC MPEG-4 AVC), including its scalable video coding (SVC) and multi-view video coding (MVC) extensions.

In addition, there is a new video coding standard, High Efficiency Video Coding (HEVC), being developed by the Joint Collaboration Team on Video Coding (JCT-VC) of the ITU-T Video Coding Experts Group (VCEG) and the ISO/IEC Moving Picture Experts Group (MPEG). The most recent working draft (WD) of HEVC is Working Draft 8, hereinafter referred to as HEVC WD8. The High Efficiency Video Coding (HEVC) Text Specification Draft 8 by Bross et al. (Stockholm, July 2012) is available from http://phenix.int-evry.fr/jct/doc_end_user/documents/10_Stockholm/wg11/JCTVC-J1003-v8.zip as of May 2, 2013.

HEVC WD8 is incorporated herein by reference in its entirety.In addition, although the techniques described in this disclosure are described with respect to the HEVC standard, aspects of this disclosure are not so limited and may be extended to other video coding standards, as well as proprietary video coding techniques.

A video encoder may generate a bitstream containing encoded video data. The bitstream may include a series of network abstraction layer (NAL) units. The NAL units of the bitstream may include video coding layer (VCL) NAL units and non-VCL NAL units. VCL NAL units may include coded slices of pictures. Non-VCL NAL units may include video parameter sets (VPSs), sequence parameter sets (SPSs), picture parameter sets (PPSs), supplemental enhancement information (SEI), or other types of data. A VPS is a syntax structure that may contain syntax elements that apply to zero or more entire coded video sequences. An SPS is a syntax structure that may contain syntax elements that apply to zero or more entire coded video sequences. A single VPS may apply to multiple SPSs. A PPS is a syntax structure that may contain syntax elements that apply to zero or more entire coded pictures. A single SPS may apply to multiple PPSs. In general, various aspects of the VPS, SPS, and PPS may be formed as defined by the HEVC standard.

A NAL unit may include a syntax element indicating the value of a temporalId variable. The temporalId of a NAL unit specifies the temporal identifier of the NAL unit. If the temporal identifier of a first NAL unit is less than the temporal identifier of a second NAL unit, the data encapsulated by the first NAL unit may be decoded without reference to the data encapsulated by the second NAL unit.

Each video coding standard typically includes specifications for a video buffering model. In AVC and HEVC, the buffering model is called a Hypothetical Reference Decoder (HRD), which describes how data is buffered for decoding and how decoded data is buffered for output. The HRD includes buffering models for both the Coded Picture Buffer (CPB) and the Decoded Picture Buffer (DPB). The CPB is a first-in, first-out buffer that contains access units in the decoding order specified by the HRD. The DPB is a buffer that holds decoded pictures for reference, output reordering, or output delay specified by the HRD. Hypothetical Reference Decoder parameters mathematically specify the behavior of the CPB and DPB. The HRD can directly impose constraints on various parameters, including timing, buffer size, and bitrate, and can indirectly impose constraints on bitstream characteristics and statistics. In some examples, a complete set of HRD parameters may include five basic parameters: initial CPB removal delay, CPB size, bitrate, initial DPB output delay, and DPB size.

In AVC and HEVC, bitstream conformance and decoder conformance are specified as part of the HRD specification. Although the name of the hypothetical reference decoder refers to a decoder, HRD is generally required on the encoder side for bitstream conformance, but not necessarily on the decoder side. However, aspects of the present invention are not so limited, and HRD can also be part of the decoder side. AVC and HEVC may specify two types of bitstream or HRD conformance, namely Type I and Type II. Furthermore, AVC and HEVC specify two types of decoder conformance: output timing decoder conformance and output order decoder conformance.

In the AVC and HEVC HRD models, decoding or CPB removal is based on access units, and the standards assume that picture decoding is instantaneous. In practice, if a conforming decoder strictly adheres to the decode time signaled (e.g., in a Supplemental Enhancement Information (SEI) message) to start decoding an access unit, then the earliest possible time to output a particular decoded picture is equal to the decode time of that particular picture plus the time required to decode that particular picture. That is, the earliest time to output a decoded picture is the decode time plus the time used to decode the picture. However, in the real world, the time required to decode a picture cannot be equal to zero.

In HEVC WD8, the Hypothetical Reference Decoder (HRD) is specified in Annex C. The HRD relies on HRD parameters, which can be provided in the bitstream in the hrd_parameters() syntax structure (in the Video Parameter Set (VPS) and/or Sequence Parameter Set (SPS)), the buffering period SEI message, and the picture timing SEI message. U.S. Provisional Application No. 61/705,102, filed on September 24, 2012, proposes enhanced signaling and selection of HRD parameters.

There may be problems associated with existing methods for signaling and derivation of CPB removal time (also referred to as decoding time). Some of these problems are described below.

When the CPB removal time for a decoding unit within an access unit depends on timing information from a previous access unit, the decoding unit CPB removal time may not be error-resilient. An access unit may include one or more decoding units. A removal time may be determined for each DU in an AU. A CPB removal time may be signaled for an AU and for one or more DUs within the AU. An SEI message for an AU may include the CPB removal time for the AU itself, which also corresponds to the CPB removal time for the last DU within the AU.

The coded picture buffer operates at two levels: the access unit level and the sub-picture level. When the CPB operates at the sub-picture level (i.e., when SubPicCpbFlag is equal to 1), the signaling and derivation of the decoding unit (DU) CPB removal time based on the picture timing SEI message may not be error-resilient if information from the previous AU in decoding order is lost. For example, the timing information signaled for the current AU includes the duration between the CPB removal times of the first DU in the current AU and the last DU in the previous AU. Therefore, if the timing information for the last DU in the previous AU is lost, the decoder cannot determine the removal time for the first DU in the current AU because the removal time for the first DU depends on the lost timing information.

In other words, the signaling of the duration between the first decoding unit in the current AU and the CPB removal time of the last DU in the previous AU in decoding order, and its use in CPB removal time derivation, makes the system or decoder vulnerable to lost timing information. For example, if the CPB removal information (i.e., picture timing SEI message) of the previous AU is lost, the CPB removal time of the first DU in the current AU cannot be correctly derived. Furthermore, except for the last DU in the current AU (whose CPB removal time is derived to be equal to the CPB removal time of the current AU), each of the CPB removal times of all other DUs in the current AU depends on the CPB removal time of the previous DU in decoding order. Therefore, if this loss occurs, the CPB removal time of every DU in the current AU (except the last DU) cannot be correctly derived.

In contrast, described herein are techniques that can reduce the vulnerability of a system or decoder to lost timing information. For example, techniques are provided for determining a coded picture buffer removal time for a DU of an AU that is independent of the removal time of any other access unit. For example, a video encoder would signal the CPB removal time for the DU of an AU in a picture timing SEI message to be received by a video decoder based on the duration between the current DU and the CPB removal time of the next DU in decoding order in the AU or the CPB removal time of the last DU in the AU. Thus, the present invention describes techniques for more error-resilient and efficient signaling and derivation of CPB removal times for coded data units in video decoding because the timing information for each DU in an AU does not depend on timing information from another different AU.

Another issue associated with existing methods for signaling and deriving CPB removal time is that timing information in sub-picture timing SEI messages may not be utilized, even if it exists. For example, a sub-picture timing SEI message may exist that carries DU CPB removal delay information. However, sub-picture-level CPB operation is specified in such a way that video decoders always utilize picture timing SEI messages and never sub-picture timing SEI messages. Therefore, bits used to represent the sub-picture timing SEI message may be wasted. Furthermore, the DU CPB removal delay signaled in the sub-picture timing SEI message is the difference between the CPB removal time of the associated DU and the CPB removal time of the first DU of the previous AU associated with the buffering period SEI message. While this may be somewhat error-resilient, it may also be inefficient because the time difference may be a significant value.

However, this disclosure provides techniques for specifying CPB removal time derivation in a manner that can utilize information carried in a sub-picture timing supplemental enhancement information (SEI) message. For example, CPB removal time derivation is specified in a manner that can utilize information carried in a sub-picture timing SEI message, and a video encoder can signal the duration between the last DU in decoding order in an AU and the CPB removal of the DU associated with the sub-picture timing SEI message, thereby making encoder signaling and decoder derivation more efficient and error resilient.

Another problem associated with existing methods for signaling and deriving CPB removal time is that sub-picture level CPB parameters for the same functionality may be present in both picture timing SEI messages and sub-picture timing SEI messages. The functionality may be provided to support sub-picture based CPB operations. Duplicating such parameters for the same functionality may be inefficient. It may be the case that only one set of sub-picture level CPB parameters in either type of SEI message is sufficient. Techniques are described herein that configure a video encoder to provide a sequence level flag that may be signaled to indicate that sub-picture CPB parameters are present in only one of the picture timing SEI message or the sub-picture timing SEI message, but not both. By using this sequence level flag, the video decoder determines whether to find sub-picture CPB parameters, e.g., sub-picture level CPB removal delay parameters, in the picture timing SEI message or in the sub-picture timing SEI message.

Another problem associated with existing methods for signaling and deriving CPB removal time is that the definition of decoding units does not take into account non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63. Therefore, when some of these non-VCL NAL units are present, unexpected sub-picture level CPB behavior may occur. In contrast, this disclosure provides techniques for extending the definition of decoding units to include non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

Another potential issue associated with existing methods for signaling and deriving CPB removal times is that buffering period SEI messages and recovery point SEI messages can be associated with AUs with any temporal identification value (TemporalId). Consequently, the encoder can initialize the HRD at AUs with a TemporalId greater than 0. In this case, when temporal scalability is supported, the CPB removal time of AUs with smaller TemporalId values within the same buffering period can depend on information in AUs with larger TemporalId values. However, for temporal scalability to work, the decoding process of any AU must not depend on another AU with a larger TemporalId. This disclosure further provides techniques for restricting buffering period SEI messages and recovery point SEI messages so that they cannot be associated with AUs with a TemporalId greater than 0.

A temporal identifier (TemporalId) is a hierarchical value that indicates which pictures can be used to code the current picture. Generally speaking, a picture with a particular TemporalId value may be a reference picture for pictures with an equal or greater TemporalId value, but not vice versa. For example, a picture with a TemporalId value of 1 may be a reference picture for pictures with TemporalId values of 1, 2, 3, etc., but not for a picture with a TemporalId value of 0.

The lowest TemporalId value can also indicate the lowest display rate. For example, if the video decoder only decodes pictures with a TemporalId value of 0, the display rate can be 7.5 pictures per second. If the video decoder only decodes pictures with TemporalId values 0 and 1, the display rate can be 15 pictures per second, and so on.

Yet another potential problem associated with existing methods for signaling and deriving CPB removal times is that, during CPB removal time derivation, when sub_pic_cpb_params_present_flag is equal to 1, the derivation of the CPB removal time uses the last arrival time and the nominal removal time for both the case where SubPicCpbFlag is equal to 0 (when the CPB operates at the AU level) and the case where SubPicCpbFlag is equal to 1 (when the CPB operates at the sub-picture level). However, those used values for the last arrival time and the nominal removal time may be derived for only one of the two cases (e.g., for SubPicCPBFlag equal to 0 or for SubPicCPBFlag equal to 1), and therefore may not be used for the other case. The techniques described herein provide a flag to signal whether the decoder will derive the CPB removal time at the AU level or the sub-picture level. For example, according to the techniques described herein, a decoder derives the CPB arrival time and nominal removal time for both the AU level and the sub-picture level, regardless of the value of SubPicCpbFlag. However, the decoder derives the CPB removal time only for the AU level when SubPicCpbFlag is 0 and only for the sub-picture level when SubPicCpbFlag is 1. As described herein, the CPB nominal removal time may be a default value for the CPB removal time. In some examples with typical conditions, the CPB removal time is equal to the CPB nominal removal time. However, under certain conditions, this may be different, and the CPB removal time may differ slightly from the default value.

The following techniques described in this disclosure may address the issues described above. For example, the techniques described in this disclosure may provide a more error-resilient determination of coded picture buffer removal time. Furthermore, in addition to improved error resilience, the techniques may also improve signaling efficiency, thereby reducing bandwidth, reducing signaling overhead, and increasing coding efficiency. Furthermore, the techniques described in this disclosure may allow for appropriate temporal scalability.

For example, such techniques may include determining a coded picture buffer removal time for a decoding unit (DU) of an access unit (AU) independent of the removal time of any other access unit (AU). For example, the CPB removal time for a DU will be signaled based on the duration between the current DU and the CPB removal time of the next DU in decoding order in the AU, or the CPB removal time of the last DU in the AU. According to the techniques described herein, the techniques may also include signaling a sequence-level flag to control the presence of sub-picture CPB parameters in only one of a picture timing SEI message or a sub-picture timing SEI message. The techniques may also include extending the definition of a decoding unit. Additional techniques provide for limiting the buffering period SEI message and the recovery point SEI message so that they cannot be associated with AUs with a variable TemporalId greater than 0. The techniques may also include providing a flag to signal whether the CPB removal time is derived at the AU level or the sub-picture level.

The implementation details for these techniques are described in more detail below. Other parts not mentioned may be the same as in HEVC WD8.

FIG1 is a block diagram illustrating an example video encoding and decoding system 10 that may utilize the techniques described in this disclosure. As shown in FIG1 , system 10 includes a source device 12 that generates encoded video data to be decoded at a later time by a destination device 14. Source device 12 and destination device 14 may comprise any of a wide range of devices, including desktop computers, notebook (e.g., laptop) computers, tablet computers, set-top boxes, telephone handsets (e.g., so-called "smart" phones), so-called "smart" pads, televisions, cameras, display devices, digital media players, video game consoles, video streaming devices, or the like. In some cases, source device 12 and destination device 14 may be equipped for wireless communication.

1 , source device 12 includes a video source 18, a video encoder 20, and an output interface 22. In some cases, output interface 22 may include a modulator/demodulator (modem) and/or a transmitter. In source device 12, video source 18 may include sources such as a video capture device (e.g., a video camera), a video archive containing previously captured video, a video feed interface that receives video from a video content provider, and/or a computer graphics system for generating computer graphics data as source video, or a combination of such sources. As an example, if video source 18 is a video camera, source device 12 and destination device 14 may form so-called camera phones or video phones. However, the techniques described in this disclosure may be applicable to video coding generally and may be applied to wireless and/or wired applications.

The captured, pre-captured, or computer-generated video may be encoded by video encoder 20. The encoded video data may be transmitted directly to destination device 14 via output interface 22 of source device 12. The encoded video data may also (or alternatively) be stored on storage device 32 for later access by destination device 14 or other devices for decoding and/or playback.

Destination device 14 includes an input interface 28, a video decoder 30, and a display device 32. In some cases, input interface 28 may include a receiver and/or a modem. Input interface 28 of destination device 14 receives encoded video data via link 16. The encoded video data communicated via link 16 or provided on storage device 32 may include various syntax elements generated by video encoder 20 for use by a video decoder, such as video decoder 30, when decoding the video data. Such syntax elements may be included with the encoded video data transmitted on a communication medium, stored on a storage medium, or stored on a file server.

Display device 32 may be integrated with destination device 14 or external to destination device 14. In some examples, destination device 14 may include an integrated display device and may also be configured to interface with an external display device. In other examples, destination device 14 may be a display device. Generally, display device 32 displays the decoded video data to a user and may include any of a variety of display devices, such as a liquid crystal display (LCD), a plasma display, a dot matrix display, an organic light emitting diode (OLED) display, electronic ink, or another type of display device.

Destination device 14 may receive the encoded video data to be decoded via link 16. Link 16 may comprise any type of medium or device capable of moving the encoded video data from source device 12 to destination device 14. In one example, link 16 may comprise a communication medium to enable source device 12 to transmit the encoded video data directly to destination device 14 in real time. The encoded video data may be modulated according to a communication standard (e.g., a wireless communication protocol) and transmitted to destination device 14. The communication medium may comprise any wireless or wired communication medium, such as a radio frequency (RF) spectrum or one or more physical transmission lines. The communication medium may form part of a packet-based network, such as a local area network, a wide area network, or a global network (e.g., the Internet). The communication medium may include routers, switches, base stations, or any other equipment that may be used to facilitate communication from source device 12 to destination device 14.

Alternatively, the encoded data may be output from output interface 22 to storage device 32. Similarly, the encoded data may be accessed from storage device 32 via input interface. Storage device 32 may include any of a variety of distributed or locally accessed data storage media, such as a hard drive, Blu-ray disc, DVD, CD-ROM, flash memory, volatile or non-volatile memory, or any other suitable digital storage medium for storing encoded video data. In yet another example, storage device 32 may correspond to a file server or another intermediate storage device that can hold the encoded video generated by source device 12. Destination device 14 may access the stored video data from storage device 32 via streaming or downloading. A file server may be any type of server capable of storing encoded video data and transmitting it to destination device 14. Example file servers include a web server (e.g., for a website), an FTP server, a network attached storage (NAS) device, or a local disk drive. Destination device 14 may access the encoded video data via any standard data connection, including an Internet connection. This may include a wireless channel (e.g., a Wi-Fi connection), a wired connection (e.g., DSL, cable modem, etc.), or a combination of both suitable for accessing encoded video data stored on a file server. The transmission of the encoded video data from storage device 32 may be a streaming transmission, a download transmission, or a combination of both.

The techniques of this disclosure are not necessarily limited to wireless applications or settings. The techniques may be applied to video coding to support any of a variety of multimedia applications, such as over-the-air television broadcasts, cable television transmissions, satellite television transmissions, streaming video transmissions (e.g., via the Internet), encoding of digital video for storage on data storage media, decoding of digital video stored on data storage media, or other applications. In some examples, system 10 may be configured to support one-way or two-way video transmission to support applications such as video streaming, video playback, video broadcasting, and/or video telephony.

Video encoder 20 and video decoder 30 may operate according to a video compression standard, such as the High Efficiency Video Coding (HEVC) standard currently under development by the Joint Collaboration Team on Video Coding (JCT-VC) of the ITU-T Video Coding Experts Group (VCEG) and the ISO/IEC Motion Picture Experts Group (MPEG). Video encoder 20 and video decoder 30 may operate according to HEVC WD8. Alternatively, video encoder 20 and video decoder 30 may operate according to other proprietary or industry standards, such as ITU-T H.261, ISO/IEC MPEG-1 Visual, ITU-T H.262 or ISO/IEC MPEG-2 Visual, ITU-T H.263, ISO/IEC MPEG-4 Visual, and ITU-T H.264 (also known as ISO/IEC MPEG-4 AVC), including its scalable video coding (SVC) and multi-view video coding (MVC) extensions, or extensions of such standards. However, the techniques of this disclosure are not limited to any particular coding standard.

In some examples, video encoder 20 and video decoder 30 may each be integrated with an audio encoder and decoder and may include appropriate MUX-DEMUX units or other hardware and software to handle the encoding of both audio and video in a common data stream or separate data streams. In some examples, the MUX-DEMUX units may conform to the ITU H.223 multiplexer protocol or other protocols such as the User Datagram Protocol (UDP), if applicable.

Video encoder 20 and video decoder 30 may each be implemented as any of a variety of suitable encoder circuits, such as one or more microprocessors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), discrete logic, software, hardware, firmware, or any combination thereof. When the techniques are partially implemented in software, the device may store instructions for the software in a suitable non-transitory computer-readable medium and execute the instructions in hardware using one or more processors to perform the techniques of this disclosure. Each of video encoder 20 and video decoder 30 may be included in one or more encoders or decoders, either of which may be integrated as part of a combined encoder/decoder (CODEC) in the respective device.

The JCT-VC is working on the development of the HEVC standard. The HEVC standardization effort is based on an evolving model of a video coding device known as the HEVC Test Model (HM). The HM assumes several additional capabilities of video coding devices compared to existing devices based on, for example, ITU-T H.264/AVC. For example, while H.264 provides nine intra-frame prediction coding modes, the HM can accommodate up to 33 intra-frame prediction coding modes.

In general, the working model of the HM describes that a video frame or picture can be divided into a series of treeblocks or largest coding units (LCUs) that include both luma and chroma samples. Treeblocks have a similar purpose to macroblocks in the H.264 standard. A slice includes many consecutive treeblocks in coding order. A video frame or picture can be partitioned into one or more slices. Each treeblock can be split into coding units (CUs) according to a quadtree. For example, a treeblock that is the root node of a quadtree can be split into four child nodes, and each child node can be a parent node and split into another four child nodes. The last unsplit child node that is a leaf node of the quadtree comprises a coding node, i.e., a coded video block. Syntax data associated with the coded bitstream can define the maximum number of times a treeblock can be split, and can also define the minimum size of a coding node.

A CU includes a coding node and a prediction unit (PU) and a transform unit (TU) associated with the coding node. The size of a CU typically corresponds to the size of the coding node and is typically square in shape. The size of a CU can range from 8x8 pixels up to the size of a treeblock having a maximum of 64x64 pixels or larger. Each CU may contain one or more PUs and one or more TUs. Syntax data associated with a CU may describe, for example, the partitioning of the CU into one or more PUs. The partitioning mode may differ depending on whether the CU is skip or direct mode coded, intra-prediction mode coded, or inter-prediction mode coded. A PU may be partitioned into a non-square shape. Syntax data associated with a CU may also describe, for example, the partitioning of the CU into one or more TUs according to a quadtree. A TU may be square or non-square in shape.

The HEVC standard allows for transforms based on TUs, which can be different for different CUs. The size of a TU is typically set based on the size of the PU within a given CU defined for a partitioned LCU, but this is not always the case. A TU is typically the same size as or smaller than a PU. In some examples, a quadtree structure called a "residual quadtree" (RQT) can be used to subdivide the residual samples corresponding to a CU into smaller units. The leaf nodes of the RQT can be referred to as transform units (TUs). Pixel difference values associated with a TU can be transformed to produce transform coefficients, which can be quantized.

In general, a PU includes data related to the prediction process. For example, when the PU is intra-mode encoded, the PU may include data describing the intra-prediction mode used for the PU. As another example, when the PU is inter-mode encoded, the PU may include data defining a motion vector for the PU. The data defining the motion vector for the PU may describe, for example, the horizontal component of the motion vector, the vertical component of the motion vector, the resolution of the motion vector (e.g., quarter-pixel precision or eighth-pixel precision), the reference picture to which the motion vector points, and/or the reference picture list for the motion vector (e.g., List 0, List 1, or List C).

In general, TUs are used for the transform and quantization process. A given CU with one or more PUs may also include one or more transform units (TUs). After prediction, video encoder 20 may calculate residual values from the video blocks identified by the coding nodes based on the PUs. The coding nodes are then updated to reference the residual values rather than the original video blocks. The residual values include pixel difference values that can be transformed into transform coefficients using the transform and other transform information specified in the TU, quantized, and scanned to produce serialized transform coefficients for entropy coding. The coding nodes may again be updated to reference these serialized transform coefficients. This disclosure generally uses the term "video block" to refer to the coding nodes of a CU. In some specific cases, this disclosure may also use the term "video block" to refer to a treeblock (i.e., an LCU or CU), which includes the coding nodes and the PUs and TUs.

A video sequence typically includes a series of video frames or pictures. A group of pictures (GOP) typically includes a series of one or more of the video pictures. A GOP may include syntax data in a header of the GOP, a header of one or more of the pictures, or elsewhere that describes the number of pictures included in the GOP. Each slice of a picture may include slice syntax data that describes the coding mode of the corresponding slice. Video encoder 20 typically operates on video blocks within individual video slices to encode the video data. A video block may correspond to a coding node within a CU. Video blocks may have a fixed or varying size and may differ in size according to a specified coding standard.

As an example, the HM supports prediction for various PU sizes. Assuming a particular CU is 2Nx2N, the HM supports intra prediction for PU sizes of 2Nx2N or NxN, and inter prediction for symmetric PU sizes of 2Nx2N, 2NxN, Nx2N, or NxN. The HM also supports asymmetric partitioning for inter prediction for PU sizes of 2NxnU, 2NxnD, nLx2N, and nRx2N. In asymmetric partitioning, the CU is unpartitioned in one direction and partitioned into 25% and 75% partitions in the other direction. The portion of the CU corresponding to the 25% partition is indicated by an "n" followed by an indication of "up," "down," "left," or "right." Thus, for example, "2NxnU" refers to a 2Nx2N CU that is partitioned horizontally into a 2Nx0.5N PU at the top and a 2Nx1.5N PU at the bottom.

In this disclosure, "NxN" and "N by N" may be used interchangeably to refer to the pixel dimensions of a video block in terms of vertical and horizontal dimensions, e.g., 16x16 pixels or 16 by 16 pixels. Generally speaking, a 16x16 block will have 16 pixels in the vertical direction (y=16) and 16 pixels in the horizontal direction (x=16). Similarly, an NxN block typically has N pixels in the vertical direction and N pixels in the horizontal direction, where N represents a non-negative integer value. The pixels in a block may be arranged in rows and columns. Furthermore, a block need not necessarily have the same number of pixels in the horizontal direction as in the vertical direction. For example, a block may comprise NxM pixels, where M is not necessarily equal to N.

After intra-predictive or inter-predictive coding of a PU of a CU, video encoder 20 may calculate residual data to which the transform specified by the TU of the CU is applied. The residual data may correspond to pixel differences between pixels of an unencoded picture and prediction values corresponding to the CU. Video encoder 20 may form the residual data for the CU and then transform the residual data to produce transform coefficients.

After performing any transforms to generate transform coefficients, video encoder 20 may perform quantization of the transform coefficients. Quantization generally refers to a process in which transform coefficients are quantized to potentially reduce the amount of data used to represent the coefficients, thereby providing further compression. The quantization process may reduce the bit depth associated with some or all of the coefficients. For example, during quantization, an n-bit value may be truncated to an m-bit value, where n is greater than m.

In some examples, video encoder 20 may utilize a predefined scan order to scan the quantized transform coefficients to produce a serialized vector that can be entropy encoded. In other examples, video encoder 20 may perform adaptive scanning. After scanning the quantized transform coefficients to form a one-dimensional vector, video encoder 20 may entropy encode the one-dimensional vector, for example, according to context-adaptive variable length coding (CAVLC), context-adaptive binary arithmetic coding (CABAC), syntax-based context-adaptive binary arithmetic coding (SBAC), probability interval partitioning entropy (PIPE) coding, or another entropy encoding method. Video encoder 20 may also entropy encode syntax elements associated with the encoded video data for use by video decoder 30 when decoding the video data.

To perform CABAC, video encoder 20 may assign a context within a context model to a symbol to be transmitted. Context may relate to, for example, whether neighboring values of a symbol are nonzero. To perform CAVLC, video encoder 20 may select a variable length code for the symbol to be transmitted. Codewords in VLC may be constructed such that relatively shorter codes correspond to more likely symbols, while longer codes correspond to less likely symbols. In this way, the use of VLC may achieve bit savings compared to, for example, using an equal-length codeword for each symbol to be transmitted. Probability determination may be based on the context assigned to the symbol.

In some examples, video encoder 20 and video decoder 30 may be configured to implement one or more example techniques described in this disclosure. Video encoder 20 may encode video data in the form of access units that are decomposed into one or more decoding units. Such access units may be temporarily stored in a coded picture buffer. Video decoder 30 may extract a DU for decoding in decoding order based on timing information included in syntax elements for the corresponding AU or DU.

According to the techniques described in this disclosure, the term "decoding unit" may be defined as follows. A decoding unit is an access unit or a subset of an access unit. If the syntax element SubPicCpbFlag is equal to 0, then the decoding unit is an access unit. Otherwise, a DU contains one or more VCL NAL units and associated non-VCL NAL units in an AU. If the non-VCL NAL unit has nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, FD_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63, then the non-VCL NAL unit is associated with the most immediately preceding VCL NAL unit in decoding order; otherwise, the non-VCL NAL unit is associated with the first subsequent VCL NAL unit in decoding order. To correctly consider non-VCL NAL units according to the techniques described herein, a DU may be defined to consider non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

According to the techniques described in this disclosure, the term "operation point" may be defined as follows. An operation point is identified by a set of nuh_reserved_zero_6bits values (denoted as OpLayerIdSet) and TemporalId values (denoted as OpTid) and an associated bitstream subset derived as the output of the sub-bitstream extraction process (as specified in subclause 10.1 of HEVC WD8). The syntax elements OpTid and OpLayerIdSet may serve as input and may be independently decodable.

Some examples of bitstreams generated according to the techniques described herein may have a level of bitstream conformance. Subclause 10.1 of HEVC WD8 describes that the requirement for bitstream conformance may be that any sub-bitstream included in the output of the process specified in subclause 10.1 that has a tIdTarget equal to any value in the range of 0 to 6 (inclusive) and a targetDecLayerIdSet containing a value of 0 may conform to HEVC.

In some examples, a conforming bitstream may contain one or more coded slice NAL units with nuh_reserved_zero_6bits equal to 0 and TemporalId equal to 0.

The input to the process described herein may be the variable tIdTarget and the list targetDecLayerIdSet. The output comprises a sub-bitstream. The sub-bitstream may be derived by removing from the bitstream all NAL units whose TemporalId is greater than tIdTarget or whose nuh_reserved_zero_6bits is not between the values in targetDecLayerIdSet.

Each NAL unit may be associated with header information. For NAL unit header semantics, the following may be specified. During decoding, decoder 30 may ignore (e.g., remove and discard from the bitstream) the contents of all NAL units that use the reserved value of nal_unit_type. In HRD operation as specified in Annex C of HEVC WD8, NAL units with the reserved value of nal_unit_type may be considered in the derivation of CPB arrival and removal times, depending on the selected operation point in the test, but may be safely ignored (removed and discarded) during decoding.

During decoding, the decoder may ignore (e.g., remove from the bitstream and discard) all NAL units with a nuh_reserved_zero_6bits value not equal to 0. In HRD operation as specified in Annex C of HEVC WD8, NAL units with a reserved value of nuh_reserved_zero_6bits may be considered in the derivation of CPB arrival and removal timing, depending on the selected operation point in the test, but may be safely ignored (removed and discarded) during decoding.

Coded picture buffer arrival and removal times may be based on two levels: the access unit level and the sub-picture level. A video coder (e.g., video encoder 20 or video decoder 30) may be configured to derive CPB arrival and CPB nominal removal times for both the access unit level and the sub-picture level, regardless of the value of the syntax element that defines whether a DU is an AU (e.g., whether the AU includes only one DU). The syntax element may be SubPicCpbFlag, which is signaled for each AU. As discussed above, when SubPicCpbFlag is equal to 0, the DU constitutes the entire AU. Otherwise, when SubPicCpbFlag is equal to a non-zero value, the DU includes one or more VCL NAL units and associated non-VCL NAL units in the AU. In some examples, the video coder may be configured to also derive CPB removal times for the AU level when the syntax element indicates that the DU is an AU. In some of these examples, the video coder may be configured to derive the CPB removal time only for the AU level when the syntax element indicates that the DU is an AU.

In some examples, a video coder (e.g., video encoder 20 or video decoder 30) may be configured to also derive the CPB removal time for the sub-picture level when the syntax element indicates that the DU is not an AU. In some of these examples, the video coder may be configured to derive the CPB removal time only for the sub-picture level when the syntax element indicates that the DU is not an AU.

The video coder may be configured to derive the CPB arrival time and the CPB nominal removal time when a second syntax element specifies that sub-picture level CPB removal delay parameters are present and the CPB may operate at either the AU level or the sub-picture level. The second syntax element may be sub_pic_cpb_params_present_flag. When sub_pic_cpb_params_present_flag is equal to 1, the sub-picture level CPB removal delay parameters are present and the CPB may operate at either the access unit level or the sub-picture level. When sub_pic_cpb_params_present_flag is equal to 0, the sub-picture level CPB removal delay parameters are not present and the CPB operates at the access unit level.

In some of the examples where sub_pic_cpb_params_present_flag is equal to 1, the video coder may be configured to set the variable subPicParamsPresentFlag equal to 0 and derive the AU initial and final arrival times. Then, the video coder may be configured to set the variable subPicParamsPresentFlag equal to 1 and derive the DU initial and final arrival times for the DUs within the AU.

Furthermore, in some examples, a video coder (e.g., video encoder 20 or video decoder 30) may be configured to code (e.g., encode or decode) the duration between the removal of the CPB of a first decoding unit in an access unit and a second DU in the access unit. In this example, the second DU follows the first DU in decoding order and is in the same AU as the first DU. The video coder may be configured to determine the removal time of the first DU based at least on the coded duration. In some techniques, the video coder may determine the removal time of the first DU without coding the initial CPB removal delay and offset. In some examples, the second DU immediately follows the first DU in the access unit. In some examples, the second DU is the last DU in the access unit in decoding order.

A video coder (e.g., video encoder 20 or video decoder 30) may also be configured to code sub-picture-level CPB parameters. In such examples, the video coder may determine a removal time for the first DU based on at least one of a coded duration and the sub-picture-level CPB parameters. For example, the sub-picture-level CPB parameters may be a sub-picture timing SEI message associated with the first DU. In some examples, the video coder may code the duration between the last DU in decoding order in the AU and the removal time of the first DU in the sub-picture SEI message. In some examples, the video coder may code a sequence-level flag to indicate the presence of the sub-picture-level CPB parameters in a picture timing SEI message or in a sub-picture timing SEI message.

For example, video encoder 20 may be configured to encode the duration between CPB removal for a first DU in an AU and a second DU in the AU. Video encoder 20 may encode sub-picture-level CPB parameters, such as sub-picture-level CPB removal delay parameters, in one of a picture timing SEI message or a sub-picture timing SEI message. Video encoder 20 may encode a flag, sub_pic_cpb_params_in_pic_timing_sei_flag, to indicate whether the sub-picture-level CPB parameters are present in the picture timing SEI message or the sub-picture timing SEI message.

For example, video decoder 30 may decode the duration between CPB removal of a first DU in an AU and a second DU in the AU. Video decoder 30 may be configured to determine the removal time of the first DU based at least on the decoded duration. In some techniques, the video coder may determine the removal time of the first DU without decoding the initial CPB removal delay and offset. Video decoder 30 may decode sub-picture level CPB parameters from a picture timing SEI message or a sub-picture timing SEI message received from video encoder 20. Video decoder 30 may determine in which SEI message to look for sub-picture level CPB parameters based on the presence of the flag sub_pic_cpb_params_in_pic_timing_sei_flag.

In some of the example techniques described in this disclosure, the temporal identification value (TemporalId) of the second DU may be no greater than the TemporalId of the first DU. In some examples, the TemporalId of the second DU may be no greater than zero.

For example, the techniques described in this disclosure may provide for a more error-resilient determination of coded picture buffer removal time. Furthermore, in addition to improved error resilience, the techniques may also facilitate signaling efficiency, thereby reducing bandwidth, reducing signaling overhead, and increasing decoding time. Furthermore, the techniques described in this disclosure may allow for improved temporal scalability.

FIG2 is a block diagram illustrating an example video encoder 20 that may implement the techniques described in this disclosure. Video encoder 20 may perform intra- and inter-coding of video blocks within a video slice. Intra-coding relies on spatial prediction to reduce or remove spatial redundancy in video within a given video frame or picture. Inter-coding relies on temporal prediction to reduce or remove temporal redundancy in video within adjacent frames or pictures of a video sequence. Intra-mode (I-mode) may refer to any of several spatial-based compression modes. Inter-mode, such as unidirectional prediction (P-mode) or bidirectional prediction (B-mode), may refer to any of several temporal-based compression modes.

2 , video encoder 20 includes a partition unit 35, a prediction processing unit 41, a summer 50, a transform processing unit 52, a quantization unit 54, and an entropy encoding unit 56. Prediction processing unit 41 includes a motion estimation unit 42, a motion compensation unit 44, and an intra-prediction processing unit 46. For video block reconstruction, video encoder 20 also includes an inverse quantization unit 58, an inverse transform processing unit 60, a summer 62, a filter unit 64, and a decoded picture buffer (DPB) 66. Decoded picture buffer 66 may also be referred to as a reference picture memory. In other examples, video encoder 20 may include more, fewer, or different functional components.

2, video encoder 20 receives video data, and partitioning unit 35 partitions the data into video blocks. This partitioning of video data may also include partitioning the video data into slices, tiles, or other larger units, as well as video block partitioning, for example, according to a quadtree structure of LCUs and CUs. Video encoder 20 generally illustrates the components that encode video blocks within a video slice to be encoded. A slice may be divided into multiple video blocks (and possibly into sets of video blocks called tiles).

Prediction processing unit 41 may select one of multiple possible coding modes (e.g., one of multiple intra-coding modes or one of multiple inter-coding modes) for the current video block based on the error results (e.g., coding rate and distortion level). Prediction processing unit 41 may provide the resulting intra- or inter-coded block to summer 50 to generate residual block data and to summer 62 to reconstruct the encoded block for use as a reference picture.

Intra-prediction processing unit 46 within prediction processing unit 41 may perform intra-predictive coding of the current video block relative to one or more neighboring blocks in the same frame or slice as the current block to be coded to provide spatial compression. Motion estimation unit 42 and motion compensation unit 44 within prediction processing unit 41 perform inter-predictive coding of the current video block relative to one or more predictive blocks in one or more reference pictures to provide temporal compression.

Motion estimation unit 42 may be configured to determine an inter-prediction mode for a video slice according to a predetermined pattern for a video sequence. The predetermined pattern may designate a video slice in the sequence as a P slice, a B slice, or a GPB slice. Motion estimation unit 42 and motion compensation unit 44 may be integrated but are illustrated separately for conceptual purposes. Motion estimation, performed by motion estimation unit 42, is the process of generating motion vectors that estimate motion for video blocks. For example, a motion vector may indicate the displacement of a PU of a video block within a current video frame or picture relative to a predictive block within a reference picture.

A predictive block is a block that is found to closely match the PU of the block to be coded, in terms of pixel difference, which may be determined by sum of absolute difference (SAD), sum of squared difference (SSD), or other difference metrics. In some examples, video encoder 20 may calculate values for sub-integer pixel positions of a reference picture stored in decoded picture buffer 66. For example, video encoder 20 may interpolate values for quarter-pixel positions, eighth-pixel positions, or other fractional pixel positions of the reference picture. Thus, motion estimation unit 42 may perform a motion search relative to full and fractional pixel positions and output a motion vector with fractional pixel precision.

Motion estimation unit 42 calculates a motion vector for a PU of a video block in an inter-coded slice by comparing the position of the PU to the position of a predictive block of a reference picture. The reference picture may be selected from a first reference picture list (List 0) or a second reference picture list (List 1), each of which identifies one or more reference pictures stored in decoded picture buffer 66. Motion estimation unit 42 sends the calculated motion vector to entropy encoding unit 56 and motion compensation unit 44.

Motion compensation performed by motion compensation unit 44 may involve retrieving or generating a predictive block based on the motion vector determined by motion estimation, possibly performing interpolation to sub-pixel precision. Upon receiving the motion vector for the PU of the current video block, motion compensation unit 44 may search one of the reference picture lists for the predictive block to which the motion vector points. Video encoder 20 forms a residual video block by subtracting pixel values of the predictive block from the pixel values of the current video block being coded, forming pixel difference values. The pixel difference values form residual data for the block and may include luma and chroma difference components. Summer 50 represents the one or more components that perform this subtraction operation. Motion compensation unit 44 may also generate syntax elements associated with the video block and video slice for use by video decoder 30 when decoding the video block of the video slice.

As an alternative to the inter-prediction performed by motion estimation unit 42 and motion compensation unit 44, as described above, intra-prediction processing unit 46 may intra-predict the current block. In particular, intra-prediction processing unit 46 may determine an intra-prediction mode to be used to encode the current block. In some examples, intra-prediction processing unit 46 may encode the current block using various intra-prediction modes (e.g., during separate encoding passes), and intra-prediction processing unit 46 (or, in some examples, mode selection unit 40) may select an appropriate intra-prediction mode to use from the tested modes. For example, intra-prediction processing unit 46 may calculate rate-distortion values using a rate-distortion analysis for the various tested intra-prediction modes and select the intra-prediction mode with the best rate-distortion characteristics among the tested modes. Rate-distortion analysis generally determines the amount of distortion (or error) between an encoded block and the original, unencoded block that was encoded to produce the encoded block, as well as the bit rate (i.e., the number of bits) used to produce the encoded block. Intra-prediction processing unit 46 may calculate ratios from the distortions and rates for the various encoded blocks to determine which intra-prediction mode exhibits the best rate-distortion value for the block.

In either case, after selecting an intra-prediction mode for a block, intra-prediction processing unit 46 may provide information indicating the selected intra-prediction mode for the block to entropy encoding unit 56. Entropy encoding unit 56 may encode the information indicating the selected intra-prediction mode according to the techniques of this disclosure. Video encoder 20 may include configuration data in the transmitted bitstream, which may include multiple intra-prediction mode index tables and multiple modified intra-prediction mode index tables (also referred to as codeword mapping tables); definitions of encoding contexts for various blocks; and indications of the most probable intra-prediction mode, intra-prediction mode index tables, and modified intra-prediction mode index tables for each of the contexts.

After prediction processing unit 41 generates a predictive block for the current video block via inter-prediction or intra-prediction, video encoder 20 forms a residual video block by subtracting the predictive block from the current video block. The residual video data in the residual block may be contained in one or more TUs and applied to transform processing unit 52. Transform processing unit 52 transforms the residual video data into residual transform coefficients using a transform, such as a discrete cosine transform (DCT) or a conceptually similar transform. Transform processing unit 52 may convert the residual video data from the pixel domain to a transform domain (e.g., the frequency domain).

Transform processing unit 52 may send the resulting transform coefficients to quantization unit 54. Quantization unit 54 quantizes the transform coefficients to further reduce the bit rate. The quantization process may reduce the bit depth associated with some or all of the coefficients. The degree of quantization may be modified by adjusting a quantization parameter. In some examples, quantization unit 54 may then perform a scan of the matrix containing the quantized transform coefficients. Alternatively, entropy encoding unit 56 may perform the scan.

After quantization, entropy encoding unit 56 may entropy encode the quantized transform coefficients. For example, entropy encoding unit 56 may perform context adaptive variable length coding (CAVLC), context adaptive binary arithmetic coding (CABAC), syntax-based context adaptive binary arithmetic coding (SBAC), probability interval partitioning entropy (PIPE) coding, or another entropy encoding method or technique. After entropy encoding by entropy encoding unit 56, the encoded bitstream may be transmitted to video decoder 30 or archived for later transmission or retrieval by video decoder 30. Entropy encoding unit 56 may also entropy encode motion vectors and other syntax elements for the current video slice being coded.

Inverse quantization unit 58 and inverse transform processing unit 60 apply inverse quantization and inverse transform, respectively, to reconstruct the residual block in the pixel domain for later use as a reference block for a reference picture. Motion compensation unit 44 may calculate a reference block by adding the residual block to a predictive block of one of the reference pictures within one of the reference picture lists. Motion compensation unit 44 may also apply one or more interpolation filters to the reconstructed residual block to calculate sub-integer pixel values for use in motion estimation. Summer 62 adds the reconstructed residual block to the motion compensated prediction block produced by motion compensation unit 44 to produce a reference block for storage in decoded picture buffer 66. The reference block may be used by motion estimation unit 42 and motion compensation unit 44 as a reference block to inter-predict a block in a subsequent video frame or picture.

Video encoder 20 also includes filter unit 64, which may filter block boundaries to remove blocking artifacts from the reconstructed video. That is, filter unit 64 may perform one or more deblocking operations to reduce blocking artifacts in the coding blocks associated with the CU. Filter unit 64 may be a deblocking filter and filters the output of summer 62. In addition to filter unit 64, additional loop filters (in-loop or post-loop) may also be used.

Decoded picture buffer 66 may store the reconstructed coding block after filter unit 64 performs one or more deblocking operations on the reconstructed coding block. Prediction processing unit 41 may use a reference picture containing the reconstructed coding block to perform inter prediction on PUs of other pictures. Furthermore, intra-prediction processing unit 46 may use the reconstructed coding block in decoded picture buffer 66 to perform intra prediction on other PUs in the same picture as the CU.

Video encoder 20 may generate syntax elements related to CPB removal time for DUs within an AU according to the techniques described herein.Once such syntax elements are generated, video encoder 20 encodes them into one or more bitstreams and outputs the bitstreams.

According to this disclosure, prediction processing unit 41 represents an example unit for performing the example functionality described above. In other examples, units other than prediction processing unit 41 may implement the examples described above. In some other examples, prediction processing unit 41, in combination with one or more other units of video encoder 20, may implement the examples described above. In still other examples, a processor or unit of video encoder 20, alone or in combination with other units of video encoder 20, may implement the examples described above.

FIG3 is a block diagram illustrating an example video decoder 30 that may implement the techniques described in this disclosure. In the example of FIG3, video decoder 30 includes an entropy decoding unit 80, a prediction processing unit 81, an inverse quantization unit 86, an inverse transform unit 88, a summer 90, and a decoded picture buffer (DPB) 92. Prediction processing unit 81 includes a motion compensation unit 82 and an intra-prediction processing unit 84. A coded picture buffer (CPB) 94 is shown as an input to video decoder 30. However, in some examples, CPB 94 may be part of video decoder 30. In some examples, video decoder 30 may perform a decoding pass that is generally reciprocal to the encoding pass described with respect to video encoder 20 from FIG2.

During the decoding process, video decoder 30 receives an encoded video bitstream representing video blocks of encoded video slices and associated syntax elements from video encoder 20. The video blocks of encoded video slices and associated syntax elements from video encoder 20 may be extracted from coded picture buffer 94. The encoded video from CPB 94 may include, for example, access units (AUs) including decoding units (DUs). Syntax elements may include variables and flags that indicate CPB removal time for access units and decoding units.

Entropy decoding unit 80 of video decoder 30 entropy decodes the bitstream to generate quantized coefficients, motion vectors, and other syntax elements. Entropy decoding unit 80 forwards the motion vectors and other syntax elements to prediction processing unit 81. Video decoder 30 may receive the syntax elements at the video slice level and/or the video block level.

When the video slice is coded as an intra-coded (I) slice, intra-prediction processing unit 84 of prediction processing unit 81 may generate prediction data for the video block of the current video slice based on the signaled intra-prediction mode and data from previously decoded blocks of the current frame or picture. When the video frame is coded as an inter-coded (i.e., B, P, or GPB) slice, motion compensation unit 82 of prediction processing unit 81 generates predictive blocks for the video block of the current video slice based on motion vectors and other syntax elements received from entropy decoding unit 80. The predictive blocks may be generated from one of the reference pictures within one of the reference picture lists. Video decoder 30 may construct reference frame lists, List 0 and List 1, using a default construction technique based on the reference pictures stored in decoded picture buffer 92.

Motion compensation unit 82 determines prediction information for video blocks of the current video slice by parsing motion vectors and other syntax elements, and uses the prediction information to generate predictive blocks for the current video block being decoded. For example, motion compensation unit 82 uses some of the received syntax elements to determine the prediction mode (e.g., intra- or inter-prediction) used to code the video blocks of the video slice, the inter-prediction slice type (e.g., B-slice, P-slice, or GPB-slice), construction information for one or more of the reference picture lists for the slice, motion vectors for each inter-coded video block of the slice, inter-prediction status for each inter-coded video block of the slice, and other information used to decode video blocks in the current video slice.

Motion compensation unit 82 may also perform interpolation based on interpolation filters. Motion compensation unit 82 may calculate interpolated values for sub-integer pixels of a reference block using interpolation filters as used by video encoder 20 during encoding of the video block. In this case, motion compensation unit 82 may determine the interpolation filters used by video encoder 20 from the received syntax elements and use the interpolation filters to produce the predictive blocks.

Inverse quantization unit 86 inverse quantizes (i.e., dequantizes) the quantized transform coefficients provided in the bitstream and decoded by entropy decoding unit 80. The inverse quantization process may include applying the quantization parameters calculated by video encoder 20 to each video block in the video slice to determine the degree of quantization, and likewise, the degree of inverse quantization that should be applied. Inverse transform processing unit 88 applies an inverse transform (e.g., an inverse DCT, an inverse integer transform, or a conceptually similar inverse transform process) to the transform coefficients to produce residual blocks in the pixel domain.

After motion compensation unit 82 generates a predictive block for the current video block based on the motion vectors and other syntax elements, video decoder 30 forms a decoded video block by summing the residual block from inverse transform processing unit 88 with the corresponding predictive block generated by motion compensation unit 82. Summer 90 represents one or more components that perform this summing operation. If desired, a deblocking filter may also be applied to filter the decoded block to remove blocking artifacts. Other loop filters (either in the decoding loop or after the decoding loop) may also be used to smooth pixel transitions or otherwise improve video quality. The decoded video blocks in a given frame or picture are then stored in DPB 92, which stores reference pictures used for subsequent motion compensation. DPB 92 also stores the decoded video for later presentation on a display device (e.g., display device 32 of FIG. 1 ).

According to this disclosure, prediction processing unit 81 represents an example unit for performing the example functionality described above. In other examples, units other than prediction processing unit 81 may implement the examples described above. In some other examples, prediction processing unit 81, in combination with one or more other units of video decoder 30, may implement the examples described above. In still other examples, a processor or unit of video decoder 30, alone or in combination with other units of video decoder 30, may implement the examples described above.

Video decoder 30 may store received video data in the form of a bitstream (including AUs and DUs) in a coded picture buffer (CPB) 94. Video decoder 30 may extract the DUs and AUs from CPB 94 at removal times determined according to syntax elements received in the bitstream by video decoder 30. Flags and variables present in SEI messages may inform video decoder 30 when to remove a DU from CPB 94. At the determined removal time of the current DU, video decoder 30 extracts the current DU from CPB 94 and decodes the DU. In some examples, video decoder 30 also extracts the AU when the current DU is the last DU of the AU.

The following describes the operation of CPB 94. This description applies independently to each of the CPB parameters present, and to the Type I and Type II conformance points shown in FIG. C-1 of HEVC WD8, where the set of CPB parameters is selected as specified in subclause C.1 of HEVC WD8. The operation of CPB 94 may include the timing of bitstream arrival and the timing of decoding unit removal and decoding of decoding units. Each is described in turn.

First, the timing of the bitstream arrival will be described. For the timing of the bitstream arrival, before the HRD is initialized, the CPB 94 is empty. In some examples, after initialization, the HRD may not be initialized again through a subsequent buffering period SEI message.

In the examples described in this disclosure, each access unit is referred to as access unit "n," where the number "n" identifies the specific access unit. The access unit associated with the buffering period SEI message that initializes CPB 94 is referred to as access unit 0. For each subsequent access unit in decoding order, the value of n is incremented by 1.

Each decoding unit is referred to as decoding unit "m," where the number "m" identifies the particular decoding unit. The first decoding unit in decoding order in access unit 0 is referred to as decoding unit 0. For each subsequent decoding unit in decoding order, the value of m is incremented by one.

When sub_pic_cpb_params_present_flag is equal to 1, for the derivation of access unit (AU) initial and final arrival times for access unit n, the following procedure is first called with the variable subPicParamsPresentFlag set to 0. Then, for the derivation of decoding unit initial and final arrival times for decoding units in access unit n, the following procedure is called with subPicParamsPresentFlag set to 1.

The variables InitCpbRemovalDelay[SchedSelIdx] and InitCpbRemovalDelayOffset[SchedSelIdx] may be set as follows: InitCpbRemovalDelay[SchedSelIdx] and InitCpbRemovalDelayOffset[SchedSelIdx] are set to the values of initial_alt_cpb_removal_delay[SchedSelIdx] and initial_alt_cpb_removal_delay_offset[SchedSelIdx], respectively, corresponding to NalHrdModeFlag in the buffering period SEI message if one of the following three conditions is true: The first condition may be that access unit 0 is a broken-link access (BLA) access unit (whose coded pictures have nal_unit_type equal to BLA_W_DLP or BLA_N_LP) and the value of rap_cpb_params_present_flag of the buffering period SEI message is equal to 1. The second condition may be DefaultInitCpbParamsFlag equal to 0. The third condition may be subPicParamsPresentFlag equal to 1. Note that in some examples, when sub_pic_cpb_params_present_flag is equal to 1, the coded video sequence may not have clear random access (CRA) or BLA pictures, and thus the first two conditions may both be false.

Otherwise, if none of the above three conditions are true, then InitCpbRemovalDelay[SchedSelIdx] and InitCpbRemovalDelayOffset[SchedSelIdx] are set to the values of initial_cpb_removal_delay[SchedSelIdx] and initial_cpb_removal_delay_offset[SchedSelIdx], respectively, corresponding to NalHrdModeFlag, in the associated buffering period SEI message selected as specified in subclause C.1 of HEVC WD8.

In the examples described herein, the time at which the first bit of decoding unit m begins to enter CPB 94 is referred to as the initial arrival time t _ai (m). The initial arrival time of decoding unit m is derived as follows. If the decoding unit is decoding unit 0 (i.e., m=0), then t _ai (0)=0. That is, the first decoding unit arrives at time 0. Otherwise, for decoding units after the first decoding unit (decoding unit m with m>0), the following applies.

If cbr_flag[SchedSelIdx] is equal to 1, then the initial arrival time of decoding unit m is equal to the last arrival time (t _ai , derived below) of decoding unit m-1 (the previous decoding unit). Equation 1 provides the relationship:

t _ai (m) = t _af (m-1) (1)

Otherwise (eg, cbr_flag[SchedSelIdx] is equal to 0), the initial arrival time of decoding unit m (eg, for m>0) is derived by equation (“Equation”) 2:

t _ai (m)＝MAX(t _af (m-1),t _ai,earliest ) (2)

The decoding unit arrival time tai _,earliest is derived as follows. If the decoding unit m is not the first decoding unit of the subsequent buffering period, then tai _,earliest is derived as shown in Equation 3:

where t _r,n (m) is the nominal removal time of decoding unit m from CPB 94 .

The final arrival time of decoding unit m is derived by Equation 4:

Where b(m) is the size of decoding unit m in bits. If the Type I conformance point applies, then b(m) contains bits that count VCL NAL units and filler data NAL units. If the Type II conformance point applies, then b(m) contains bits that count all bits of the Type II bitstream for the Type II conformance point. Type I and Type II conformance points are as shown in Figure C-1 of Annex C of HEVC WD8.

The values of SchedSelIdx, BitRate[SchedSelIdx], and CpbSize[SchedSelIdx] are constrained as follows: If the contents of the selected hrd_parameters() syntax structure for the access unit containing decoding unit m differ from those of the previous AU (in decoding order) in decoding order, the delivery scheduler (HSS) selects a value of SchedSelIdx, SchedSelIdx1, from the values of SchedSelIdx provided in the selected hrd_parameters() syntax structure for the access unit containing decoding unit m that results in BitRate[SchedSelIdx1] or CpbSize[SchedSelIdx1] for the access unit containing decoding unit m. The value of BitRate[SchedSelIdx1] or CpbSize[SchedSelIdx1] may be different from the value of BitRate[SchedSelIdx0] or CpbSize[SchedSelIdx0] for the previous access unit SchedSelIdx0. Otherwise, if the contents of the selected hrd_parameters() syntax structure are the same for both AUs, the HSS continues to operate with the previous values of SchedSelIdx, BitRate[SchedSelIdx], and CpbSize[SchedSelIdx].

When the HSS selects a value for BitRate[SchedSelIdx] or CpbSize[SchedSelIdx] that is different from the value of BitRate[SchedSelIxd] or CpbSize[SchedSelIdx] of the previous access unit, the following applies. The variable BitRate[SchedSelIdx] becomes effective at time t _ai (m). The variable CpbSize[SchedSelIdx] becomes effective under certain conditions.

If the new value of CpbSize[SchedSelIdx] is greater than the old CPB size, then CpbSize[SchedSelIdx] takes effect at time t _ai (m). Otherwise, if the new value of CpbSize[SchedSelIdx] is less than or equal to the old CPB size, then the new value of CpbSize[SchedSelIdx] takes effect at the CPB removal time of the last decoding unit of the access unit containing decoding unit m.

When SubPicCpbFlag is 1, the initial CPB arrival time t _ai (n) of access unit n is set to the initial CPB arrival time of the first decoding unit in access unit n. The final CPB arrival time t _af (n) of access unit n is set to the final CPB arrival time of the last decoding unit in access unit n. When SubPicCpbFlag is 0, each DU is an AU; therefore, the initial and final CPB arrival times of access unit n are the initial and final CPB arrival times of decoding unit n.

This disclosure now turns to describing the operation of CPB 94 with respect to the timing of decoding unit removal and decoding of decoding units.

The variables InitCpbRemovalDelay[SchedSelIdx] and InitCpbRemovalDelayOffset[SchedSelIdx] are used for DU removal time. These two variables are set as follows. If either of the two conditions is true, InitCpbRemovalDelay[SchedSelIdx] and InitCpbRemovalDelayOffset[SchedSelIdx] are set to the values of Initial_alt_cpb_removal_delay[SchedSelIdx] and Initial_alt_cpb_removal_delay_offset[SchedSelIdx], respectively, corresponding to NalHrdModeFlag, in the buffering period SEI message. The first condition is that access unit 0 is a BLA access unit (its coded picture has nal_unit_type equal to BLA_W_DLP or BLA_N_LP), and the value of rap_cpb_params_present_flag in the buffering period SEI message is equal to 1. The second condition is that DefaultInitCpbParamsFlag is equal to 0

If neither of those two conditions is true, then InitCpbRemovalDelay[SchedSelIdx] and InitCpbRemovalDelayOffset[SchedSelIdx] are set to the values of initial_cpb_removal_delay[SchedSelIdx] and initial_cpb_removal_delay_offset[SchedSelIdx], respectively, corresponding to NalHrdModeFlag, in the associated buffering period SEI message selected as specified in subclause C.1 of Annex C of HEVC WD8.

When sub_pic_cpb_params_present_flag is equal to 1, the variable CpbRemovalDelay(m), which is related to the delay time for removing decoding unit m from CPB 94, may be derived as follows. If sub_pic_cpb_params_in_pic_timing_sei_flag is equal to 0, then CpbRemovalDelay(m) is set to du_spt_cpb_removal_delay in the sub-picture timing SEI message associated with decoding unit m. The sub-picture timing SEI message may be selected as specified in subclause C.1 of Annex C of HEVC WD8.

If du_common_cpb_removal_delay_flag is equal to 0, then the variable CpbRemovalDelay(m) is set to the value of du_cpb_removal_delay_minus1[i] + 1 for decoding unit m in the picture timing SEI message selected as specified in subclause C.1 of Annex C of HEVC WD8, associated with the access unit containing decoding unit m. The value of i is 0 for the first num_nalus_in_du_minus1[0]+1 consecutive decoding units in the access unit containing decoding unit m, 1 for the following num_nalus_in_du_minus1[1]+1 decoding units in the same access unit, 2 for the following num_nalus_in_du_minus1[2]+1 decoding units in the same access unit, and so on.

Otherwise, if sub_pic_cpb_params_present_flag is not equal to 1 and du_common_cpb_removal_delay_flag is not equal to 0, then CpbRemovalDelay(m) is set to the value of du_common_cpb_removal_delay_minus1+1 in the picture timing SEI message selected as specified in subclause C.1 of Annex C of HEVC WD8, associated with the access unit containing decoding unit m.

The nominal removal time of access unit n from CPB 94 may also be determined as follows. If access unit n is access unit 0 (i.e., the access unit that initializes the HRD), then the nominal removal time tr _,n (0) of access unit 0 from CPB 94 is specified by Equation 5:

Otherwise, for access unit n (where n is non-zero or the HRD has not been initialized), the following applies. When access unit n is the first access unit of a buffering period without an initialized HRD, then the nominal removal time _tr,n (n) of access unit n from CPB 94 is given by Equation 6:

t _r,n (n)＝t _r,n (n _b )+t _c ·(au_cpd_removal_delay_minus1(n)+1) (6)

where t _r,n (n _b ) is the nominal removal time of the first access unit of the previous buffering period, and au_cpd_removal_delay_minus1(n) is the value of au_cpd_removal_delay_plus1 in the picture timing SEI message associated with access unit n, selected as specified in HEVC WD8 Annex C, subclause C.1. When access unit n is the first access unit of a buffering period, at the nominal removal time t _r,n (n) of access unit n, set n _b equal to n. When access unit n is not the first access unit of a buffering period, t _r,n (n) is given by Equation 6, where t _r,n (n _b ) is the nominal removal time of the first access unit of the current buffering period.

When sub_pic_cpb_params_present_flag is equal to 1, the nominal removal time for removing decoding unit m from CPB 94 is specified as follows, where t _r,n (n) is the nominal removal time of access unit n: If decoding unit m is the last decoding unit in access unit n, then the nominal removal time t _r,n (m) for decoding unit m is set to t _r,n (n). That is, the access unit and its last decoding unit are removed from CPB 94 at approximately the same time. Otherwise (that is, decoding unit m is not the last decoding unit in access unit n), the nominal removal time t _r,n (m) for decoding unit m is derived as shown in Equation 7, where t _r,n (n) is the nominal removal time of access unit n.

The removal time of access unit n from CPB 94 is specified as follows in Equation 8, where _taf (m) and tr _,n (m) are the final arrival time and nominal removal time of the last decoding unit in access unit n, respectively.

When SubPicCpbFlag is equal to 1, the removal time of decoding unit m from CPB 94 is specified as follows: If low_delay_hrd_flag is equal to 0 or t _r,n (m) ≥ t _af (m), then the removal time of decoding unit m is specified by Equation 9:

t _r (m)＝t _r,n (m) (9)

Otherwise, if decoding unit m is not the last decoding unit of access unit n, then the removal time of decoding unit m is specified by Equation 10:

Otherwise, if decoding unit m is the last decoding unit of access unit n, then the removal time of decoding unit m is specified by Equation 11:

t _r (m)＝t _r,n (n) (11)

In some examples, when low_delay_hrd_flag is equal to 1 and t _r,n (m) < t _af (m), the size b(m) of decoding unit m is so large that it prevents it from being removed at the nominal removal time.

At the CPB removal time of decoding unit m, decoding unit is instantaneously decoded. After the last decoding unit of picture n is decoded, the picture is considered decoded.

The following tables illustrate syntax and semantics that may be used to implement example techniques described in this disclosure. Table 1 provides example syntax and semantics for a buffering period SEI message. Table 2 provides example syntax and semantics for a picture timing SEI message. The functionality of CPB 94 may be determined by the syntax and semantics of the SEI messages. For example, video decoder 30 extracts DUs from CPB 94 based at least in part on the buffering period and picture timing SEI messages.

The Buffering Period Supplemental Enhancement Information (SEI) message provides information about the initial CPB removal delay and the initial CPB removal delay offset. The Buffering Period SEI message syntax may be the same as the Buffering Period SEI message syntax described in U.S. Provisional Application No. 61/705,102, filed on September 24, 2012, with the following semantic changes. The Buffering Period SEI message syntax is provided in Table 1 below.

Table 1: Buffering period SEI message syntax

The buffering period is specified as the set of access units between two consecutive instances of the buffering period SEI message in decoding order.

The following applies to the buffering period SEI message syntax and semantics.A bitstream (or a portion thereof) refers to a bitstream subset (or a portion thereof) associated with any of the operation points to which the buffering period SEI message applies.

For the buffering period SEI message, the syntax elements initial_cpb_removal_delay_length_minus1 and sub_pic_cpb_params_present_flag, and the variables NalHrdBpPresentFlag, VclHrdBpPresentFlag, CpbSize[SchedSelIdx], BitRate[SchedSelIdx], and CpbCnt may be found in or derived from syntax elements in the hrd_parameters() syntax structure and the sub_layer_hrd_parameters() syntax structure that may be applicable to any of the operation points to which the buffering period SEI message applies.

A buffering period SEI message may have two operation points with different OpTid values, tIdA and tIdB. Having any two operation points with different OpTid values indicates that the values of cpb_cnt_minus1[tIdA] and cpb_cnt_minus1[tIdB] coded in the hrd_parameters() syntax structure applicable to the respective operation points are the same. Additionally, a buffering period SEI message may have two operation points in the buffering period SEI message with different OpLayerIdSet values, layerIdSetA and layerIdSetB. Having any two operation points with different OpLayerIdSet values indicates that the values of nal_hrd_parameters_present_flag and vcl_hrd_parameters_present_flag are the same for the two hrd_parameters() syntax structures applicable to the two operation points, respectively.

If NalHrdBpPresentFlag or VclHrdBpPresentFlag is equal to 1, the buffering period SEI message applicable to the specified operation point may be present in any AU in the coded video sequence with a TemporalId equal to 0, and the buffering period SEI message applicable to the specified operation point may be present in each random access point (RAP) AU and in each AU associated with the recovery point SEI message. Otherwise (both NalHrdBpPresentFlag and VclHrdBpPresentFlag are equal to 0), no access unit in the coded video sequence may have a buffering period SEI message applicable to the specified operation point.

For some applications, frequent presence of buffering period SEI messages may be desirable.

When a SEI NAL unit containing a buffering period SEI message and with nuh_reserved_zero_6bits equal to 0 is present, the SEI NAL unit may precede the first VCL NAL unit in the AU in decoding order.

The access unit associated with the buffering period SEI message may have a TemporalId equal to 0.

The variable CpbCnt is derived to be equal to cpb_cnt_minus1[tId]+1, where cpb_cnt_minus1[tId] is coded in the hrd_parameters() syntax structure applicable to any of the operation points to which the buffering period SEI message applies and with OpTid equal to tId.

The following syntax elements and variables in the buffering period SEI message may be defined as follows: seq_parameter_set_id refers to the active sequence parameter set. The value of seq_parameter_set_id may be equal to the value of seq_parameter_set_id in the picture parameter set ("PPS") referenced by the coded picture associated with the buffering period SEI message. The value of seq_parameter_set_id may be in the range of 0 to 31, inclusive.

The flag rap_cpb_params_present_flag equal to 1 specifies the presence of the initial_alt_cpb_removal_delay[SchedSelIdx] and initial_alt_cpb_removal_delay_offset[SchedSelIdx] syntax elements. When not present, the value of rap_cpb_params_present_flag may be inferred to be equal to 0. The value of rap_cpb_params_present_flag may be equal to 0 when the associated picture is neither a CRA picture nor a BLA picture.

The sequence elements initial_cpb_removal_delay[SchedSelIdx] and initial_alt_cpb_removal_delay[SchedSelIdx] specify the default and alternative initial CPB removal delays for the SchedSelIdx-th CPB, respectively. The syntax elements have a length in bits given by initial_cpb_removal_delay_length_minus1+1 and are in units of, for example, a 90 kHz clock. The value of the syntax element may not be equal to 0 and may be less than or equal to

It is the time equivalent of the CPB size in units of a 90kHz clock.

The syntax elements initial_cpb_removal_delay_offset[SchedSelIdx] and initial_alt_cpb_removal_delay_offset[SchedSelIdx] specify the default and alternative initial CPB removal offsets for the SchedSelIdx-th CPB, respectively. These syntax elements have a length (in bits) given by initial_cpb_removal_delay_length_minus1+1 and are in units of a 90 kHz clock. These syntax elements may not be used by the decoder and may only be needed for the delivery scheduler (HSS) specified in Annex C of HEVC WD8.

Over the entire coded video sequence, the sum of initial_cpb_removal_delay[SchedSelIdx] and initial_cpb_removal_delay_offset[SchedSelIdx] may be constant for each value of SchedSelIdx, and the sum of initial_alt_cpb_removal_delay[SchedSelIdx] and initial_alt_cpb_removal_delay_offset[SchedSelIdx] may be constant for each value of SchedSelIdx.

The picture timing SEI message provides information on the CPB removal delay and DPB output delay of the access unit associated with the SEI message. An example of the picture timing SEI message syntax and semantics is shown in Table 2 below.

Table 2: Picture Timing SEI Message Syntax and Semantics

The following applies to picture timing SEI message syntax and semantics. The syntax elements sub_pic_cpb_params_present_flag, sub_pic_cpb_params_in_pic_timing_sei_flag, cpb_removal_delay_length_minus1, dpb_output_delay_length_minus1, and du_cpb_removal_delay_length_minus1, and the variable CpbDpbDelaysPresentFlag may be found in or derived from syntax elements in the hrd_parameters() and sub_layer_hrd_parameters() syntax structures that may be applicable to any of the operation points to which the picture timing SEI message applies.

A bitstream (or a portion thereof) refers to a bitstream subset (or a portion thereof) associated with any of the operation points to which the picture timing SEI message applies. However, note that the syntax of the picture timing SEI message may depend on the contents of the hrd_parameters() syntax structure applicable to the operation point to which the picture timing SEI message applies. Such hrd_parameters() syntax structures may be in the active video parameter set and/or sequence parameter set for the coded picture associated with the picture timing SEI message. When a picture timing SEI message is associated with a CRA access unit, an IDR access unit, or a BLA access unit that is the first access unit in the bitstream, activation of the video parameter set and sequence parameter set (and, for an IDR or BLA picture that is not the first picture in the bitstream, determination that the coded picture is an IDR or BLA picture) does not occur until the first coded slice NAL unit of the coded picture is decoded, unless it is preceded by a buffering period SEI message within the same access unit. Since the coded slice NAL units of a coded picture follow the picture timing SEI message in NAL unit order, there may be situations where it is necessary for the decoder to store the original byte sequence payload (RBSP) containing the picture timing SEI message until the active video parameter set and/or active sequence parameter set is determined and then perform parsing of the picture timing SEI message.

The presence of picture timing SEI messages in the bitstream is specified as follows. If CpbDpbDelaysPresentFlag is equal to 1, then one picture timing SEI message applicable for the specified operation point may be present in each access unit of the coded video sequence. Otherwise, for example, if CpbDpbDelaysPresentFlag is equal to 0, then no picture timing SEI message applicable for the specified operation point may be present in any access unit of the coded video sequence.

When a SEI NAL unit containing a picture timing SEI message and with nuh_reserved_zero_6bits equal to 0 is present, the SEI NAL unit may precede the first VCL NAL unit in an access unit in decoding order.

The syntax element au_cpb_removal_delay_minus1 plus 1 specifies how many clock ticks to wait after removing the access unit associated with the most recent buffering period SEI message in the previous access unit from the CPB (e.g., CPB 94) before removing the access unit associated with the picture timing SEI message from CPB 94. This value can also be used to calculate the earliest possible arrival time of access unit data at the CPB for the HSS. The syntax element is a fixed-length code whose length (in bits) is given by cpb_removal_delay_length_minus1+1.

The value of cpb_removal_delay_length_minus1, which determines the length (in bits) of the syntax element au_cpb_removal_delay_minus1, is the value of cpb_removal_delay_length_minus1 coded in the active video parameter set or sequence parameter set for the coded picture associated with the picture timing SEI message, although au_cpb_removal_delay_minus1 plus 1 specifies the number of clock ticks relative to the removal time of the previous access unit containing the buffering period SEI message (which may be an access unit of a different coded video sequence).

The syntax element pic_dpb_output_delay is used to calculate the DPB output time of a picture. pic_dpb_output_delay specifies how many clock ticks to wait before outputting a decoded picture from the DPB after the last decoding unit in an access unit is removed from the CPB. A picture may not be removed from the DPB at its output time while it is still marked as "used for short-term reference" or "used for long-term reference." In some examples, only one pic_dpb_output_delay is specified for a decoded picture.

The length of the syntax element pic_dpb_output_delay is given (in bits) by dpb_output_delay_length_minus1+1. When sps_max_dec_pic_buffering[minTid] is equal to 1, where minTid is the minimum of the OpTid values of all operation points to which the picture timing SEI message applies, pic_dpb_output_delay shall be equal to 0. The output time derived from pic_dpb_output_delay for any picture output from a decoder that conforms to the output timing may precede the output time derived from pic_dpb_output_delay for all pictures in any subsequently coded video sequence in decoding order. The picture output order established by the value of this syntax element shall be the same order as the order established by the value of PicOrderCntVal (i.e., the POC value indicating the output or display order of the pictures).

For pictures that are not output by the "promotion" process because they precede, in decoding order, an instantaneous decoding refresh (IDR) picture or a broken-link access (BLA) picture with no_output_of_prior_pics_flag equal to 1 or inferred to be equal to 1, the output time derived from pic_dpb_output_delay may increase as the value of PicOrderCntVal increases for all pictures within the same coded video sequence.

The syntax element num_decoding_units_minus1 plus 1 specifies the number of decoding units in the access unit associated with the picture timing SEI message. The value of num_decoding_units_minus1 can be in the range of 0 to PicSizeInCtbsY-1, inclusive. The flag du_common_cpb_removal_delay_flag equal to 1 specifies that the syntax element du_common_cpb_removal_delay_minus1 is present. du_common_cpb_removal_delay_flag equal to 0 specifies that the syntax element du_common_cpb_removal_delay_minus1 is not present.

The syntax element du_common_cpb_removal_delay_minus1 plus 1 specifies the duration, in sub-picture clock ticks (see subclause E.2.1 of HEVC WD8), between the removal of any two consecutive decoding units in decoding order from the CPB (e.g., CPB 94) in the access unit associated with the picture timing SEI message. This value is also used to calculate the earliest possible arrival time of decoding unit data at the CPB for the HSS, as specified in HEVC WD8 Annex C. The syntax element is a fixed-length code whose length (in bits) is given by du_cpb_removal_delay_length_minus1+1.

The syntax element num_nalus_in_du_minus1[i] plus 1 specifies the number of NAL units in the i-th DU of the AU associated with the picture timing SEI message. The value of num_nalus_in_du_minus1[i] should be in the range of 0 to PicSizeInCtbsY-1, inclusive. For example, video decoder 30 may determine how many NAL units are in the current DU based on decoding the syntax element num_nalus_in_du_minus1[i] plus 1 from the picture timing SEI message.

The first DU of an AU may include the first num_nalus_in_du_minus1[0]+1 consecutive NAL units in the AU in decoding order. The i-th (where i is greater than 0) DU of an AU consists of the num_nalus_in_du_minus1[i]+1 consecutive NAL units that immediately follow the last NAL unit in the previous DU of the AU in decoding order. At least one VCL NAL unit may be present in each DU. All non-VCL NAL units associated with a VCL NAL unit should be included in the same DU as the VCL NAL unit. Video decoder 30 determines the NAL units in a DU based on decoding syntax elements such as num_nalus_in_du_minus1[i].

The syntax element du_cpb_removal_delay_minus1[i] plus 1 specifies the duration, in sub-picture clock ticks, between the (i+1)th DU in decoding order in the AU associated with the picture timing SEI message and the removal of the i-th DU from the CPB (e.g., CPB 94). This value can also be used to calculate the earliest possible arrival time of DU data at the CPB for the HSS, as specified in HEVC WD8 Annex C. The syntax element is a fixed-length code whose length (in bits) is given by du_cpb_removal_delay_length_minus1+1.

In some examples, although the length of the syntax element is the same as du_common_cpb_removal_delay_minus1, the value may have been specified relative to the CPB removal time of the AU at the start of the buffering period. For example, video decoder 30 may determine the value of the syntax element relative to the decoded CPB removal time. In some examples, this may not be consistent with the semantics of du_common_cpb_removal_delay_minus1. For example, this may potentially conflict with Equation 7 (Equation C-10 in Appendix C of HEVC WD8), which defines that if sub_pic_cpb_params_in_pic_timing_sei_flag is indicated, then t _r,n (m) = t _r,n (m+1) - t _{c_sub} · CpdRemovaIDelay(m).

In some examples, du_cpb_removal_delay_minus1[i] plus 1 instead specifies the duration, in sub-picture clock ticks, between the removal of the AU associated with the picture timing SEI message from the CPB and the i-th DU in the AU associated with the picture timing SEI message. In this case, signaling the value of the last DU in the AU can be avoided. Therefore, video decoder 30 does not need to determine the value of the last DU in the AU from the picture timing SEI message because the removal time of the last DU is the same as the removal time of the corresponding AU.

Alternatively, in the semantics of au_cpb_removal_delay_minus1, du_common_cpb_removal_delay_minus1 and du_cpb_removal_delay_minus1[i] specify the delay/difference/duration between the “nominal CPB removal time” instead of the “CPB removal time”.

Table 3 below provides an example sub-picture timing SEI message syntax. The sub-picture timing SEI message provides CPB removal delay information for the decoding unit associated with the SEI message. An example sub-picture timing SEI message syntax and semantics are as follows.

Table 3: Sub-picture timing SEI message syntax

The following applies to sub-picture timing SEI message syntax and semantics. The syntax elements sub_pic_cpb_params_present_flag, sub_pic_cpb_params_in_pic_timing_sei_flag, and cpb_removal_delay_length_minus1, and the variable CpbDpbDelaysPresentFlag may be found in or derived from syntax elements in the hrd_parameters() and sub_layer_hrd_parameters() syntax structures applicable to any of the operation points to which the sub-picture timing SEI message applies. A bitstream (or a portion thereof) refers to a bitstream subset (or a portion thereof) associated with any of the operation points to which the sub-picture timing SEI message applies.

The presence of a sub-picture timing SEI message in the bitstream is specified as follows. If CpbDpbDelaysPresentFlag is equal to 1, sub_pic_cpb_params_present_flag is equal to 1, and sub_pic_cpb_params_in_pic_timing_sei_flag is equal to 0, then there may be one sub-picture timing SEI message applicable for the specified operation point in each decoding unit in the coded video sequence. Otherwise, no sub-picture timing SEI message applicable for the specified operation point is present in the coded video sequence. Therefore, if video decoder 30 decodes the flag and determines that the value is set as above, video decoder 30 determines that there is no sub-picture timing SEI message applicable for the specified operation point.

A decoding unit associated with a sub-picture timing SEI message consists, in decoding order, of the SEI NAL unit containing the sub-picture timing SEI message, followed by one or more NAL units that do not contain the sub-picture timing SEI message (including all subsequent NAL units in the AU up to, but not including, any subsequent SEI NAL unit containing the sub-picture timing SEI message). At least one VCL NAL unit may be present in each DU. All non-VCL NAL units associated with a VCL NAL unit may be included in the same DU as the VCL NAL unit.

In some examples, the syntax element du_spt_cpb_removal_delay specifies the duration, in units of sub-picture clock ticks (see subclause E.2.1 of HEVC WD8), between the removal from the CPB of the last decoding unit in decoding order in the current access unit containing the sub-picture timing SEI message and the decoding unit associated with the sub-picture timing SEI message. This value can also be used to calculate the earliest possible arrival time of decoding unit data at the CPB for the HSS, as specified in HEVC WD8 Annex C. The syntax element is represented by a fixed-length code whose length (in bits) is given by du_cpb_removal_delay_length_minus1+1. When the DU associated with the sub-picture timing SEI message is the last DU in the current AU, the value of du_spt_cpb_removal_delay shall be equal to 0.

Alternatively, in other examples, the syntax element du_spt_cpb_removal_delay specifies the duration, in units of sub-picture clock ticks (see subclause E.2.1 of HEVC WD8), between the removal of the next DU in decoding order in the current AU containing the sub-picture timing SEI message from the CPB 94 and the DU associated with the sub-picture timing SEI message. This value can also be used to calculate the earliest possible arrival time of decoding unit data at the CPB 94 for the HSS, as specified in HEVC WD8 Annex C. The syntax element is represented by a fixed-length code whose length (in bits) is given by du_cpb_removal_delay_length_minus1+1. When the decoding unit associated with the sub-picture timing SEI message is the last decoding unit in the current access unit, the value of du_spt_cpb_removal_delay should be equal to 0. Alternatively, no sub-picture timing SEI message is associated with the last decoding unit in each access unit.

In some examples, the syntax element du_spt_cpb_removal_delay is alternatively coded as du_spt_cpb_removal_delay_minus1. The syntax element du_spt_cpb_removal_delay_minus1 plus 1 specifies how many sub-picture clock ticks to wait before removing the DU associated with the sub-picture timing SEI message from the CPB 94 after the video decoder 30 removes the last DU in the AU associated with the most recent buffering period SEI message for the previous AU from the CPB 94. This value may also be used to calculate the earliest possible arrival time of decoding unit data at the CPB for the HSS, as specified in HEVC WD8 Annex C. The syntax element is represented by a fixed-length code whose length (in bits) is given by cpb_removal_delay_length_minus1+1.

Table 4 provided below describes an example of HRD parameter syntax and semantics. For the following syntax elements whose semantics are not included, their semantics are the same as those in U.S. Provisional Application No. 61/705,102 filed on September 24, 2012. The HRD parameter syntax and semantics may be as follows.

Table 4: HRD parameter syntax and semantics

The syntax element sub_pic_cpb_params_in_pic_timing_sei_flag equal to 1 specifies that the sub-picture level CPB removal delay parameters are present in the picture timing SEI message and that no sub-picture timing SEI message is present. The syntax element sub_pic_cpb_params_in_pic_timing_sei_flag equal to 0 specifies that the sub-picture level CPB removal delay parameters are present in the sub-picture timing SEI message and that the picture timing SEI message does not include the sub-picture level CPB removal delay parameters.

The syntax element sub_pic_cpb_params_present_flag equal to 1 specifies that the sub-picture level CPB removal delay parameters are present and the CPB can operate at either the access unit level or the sub-picture level. When sub_pic_cpb_params_present_flag is equal to 0, it specifies that the sub-picture level CPB removal delay parameters are not present and the CPB operates at the access unit level. When sub_pic_cpb_params_present_flag is not present, its value can be inferred to be equal to 0.

FIG4 is a conceptual diagram illustrating two access units 100 and 102 in consecutive decoding order that may have decoding times according to the techniques described in this disclosure. Example coded picture buffer removal times will be discussed with respect to the AUs 100 and 102 and syntax elements and variables used for SEI messages associated with the AUs 100 and 102. FIG4 also illustrates a timeline 130.

As described herein, AU 100 is access unit n, and AU 102 is access unit n+1, where n is earlier in time than n+1 in decoding order. AU 100 includes four decoding units 110-1, 110-2, 110-3, and 110-4 (collectively referred to as "decoding units 110"). As described herein, for example, DU 110-1 may be referred to as DU-M, DU 110-2 may be referred to as DU-M+1, DU 110-3 may be referred to as DU-M+2, and DU 110-4 may be referred to as DU-M+3. AU 102 includes four decoding units 112-1, 112-2, 112-3, and 112-4 (collectively referred to as "decoding units 112").

Similarly, as described herein, for example, DU 112-1 may be referred to as DU-M, DU 112-2 may be referred to as DU-M+1, DU 112-3 may be referred to as DU-M+2, and DU 112-4 may be referred to as DU-M+3. However, any access unit may be "access unit n," and any decoding unit may be "decoding unit m." In other examples, AUs 100 and 102 may have different numbers of DUs 110 and 112, respectively. Any DU 110 or 112 may be a non-video coding layer (VCL) network abstraction layer (NAL) unit with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

In this example, access units 100 and 102 are stored in a coded picture buffer (e.g., CPB 94 of FIG. 3 ). Video decoder 30 extracts decoding units 110 and 112 and access units 100 and 102 from CPB 94 for decoding at a determined time. The time used to extract an AU or DU from CPB 94 is referred to as a CPB removal time. As shown in FIG. 4 , the CPB removal times for DU 110 in AU 100 are CPB removal times 140-1, 140-2, 140-3, and 140-4 (collectively, “CPB removal times 140”). Similarly, the CPB removal times for DU 112 in AU 102 are CPB removal times 142-1, 142-2, 142-3, and 142-4 (collectively, “CPB removal times 142”). The CPB removal time of an AU may be the same as the CPB removal time of the last DU of the AU. For example, the CPB removal time of the AU 100 is substantially the same as the CPB removal time of the DU 110 - 4 (CPB removal time 140 - 4 ).

In one example, for each DU 112 in an AU 102, the duration between the next DU 112 in decoding order in the AU 102 and the CPB removal time of the particular DU 112 is signaled. For example, DU 112-2 is the current DU to be extracted from CPB 94 in decoding order and decoded by video decoder 30. The duration 132 between the CPB removal time 142-2 for DU 112-2 and the CPB removal time 142-3 for DU 112-3 (the next DU in decoding order) is signaled, for example, in an SEI message associated with access unit 102. Video decoder 30 determines the CPB removal time 142-2 for DU 112-2 based on the signaled duration 132. That is, video decoder 30 may derive the CPB removal time for each DU 112 in access unit 102 based on the removal time for DU 112 within AU 102 and not based on any removal times for other DUs within other AUs (e.g., previous AU 100 in decoding order). Thus, video decoder 30 may have improved signaling and error resilience for CPB removal times of DUs and AUs.

The CPB removal time for DU 112-2 of AU 102 may be signaled in an alternative manner. For example, the duration 134 between the CPB removal time 142-2 for DU 112-2 and the CPB removal time 142-4 for the last DU (DU 112-4) in AU 102 is signaled in an SEI message associated with AU 102. Video decoder 30 determines the CPB removal time 142-2 for DU 112-2 based on the signaled CPB removal time 142-4 of DU 112-4.

In any of the above examples, video decoder 30 determines the CPB removal time for a DU based on other DUs within the same AU. In this way, the CPB removal time for any DU does not depend on any other AUs besides the AU for that particular DU. Loss of CPB removal timing information in a previous AU will not result in an incorrect derivation of the CPB removal time for the current AU. For example, loss of CPB removal time 140 for AU 100 will not affect the determination of CPB removal time 142 for AU 102. Thus, video decoder 30 can have improved signaling and error resilience for determining CPB removal times for DUs and AUs.

The video decoder 30 may also determine the CPB removal time based at least in part on the sub-picture level CPB parameters carried in the sub-picture timing SEI message. In this example, the signaling of the CPB removal time and the derivation of the CPB removal time are both efficient and error resilient. A sequence level flag may be signaled to control the presence of the sub-picture level CPB parameters in the picture timing SEI message or in the sub-picture timing SEI message, but never in both. The sequence level flag may be the sub_pic_cpb_params_present_flag described above. The flag may also control which type of SEI message the sub-picture level CPB parameters from are used for the sub-picture level CPB operation. When sub_pic_cpb_params_present_flag is equal to 1, the CPB arrival time and the CPB removal time are signaled for both the AU level and the sub-picture level, regardless of the value of SubPicCpbFlag.

In some examples, if an AU (eg, AU 100 ) has a TemporalId greater than 0, no buffering period SEI message or recovery point SEI message may be associated with AU 100 .

FIG5 is a flowchart illustrating a method for determining a coded picture buffer (CPB) removal time for a first decoding unit in an access unit based on a CPB removal time for a second decoding unit of the access unit, according to the techniques described in this disclosure. A video decoder may perform the method of FIG5 . For example, the video decoder may be video decoder 30 of FIG1 or FIG3 .

5 includes decoding a duration between CPB removal of a first DU in an AU and CPB removal of a second DU, wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU (200). The second DU may immediately follow the first DU in the AU in decoding order. Alternatively, the second DU may be the last DU in the AU in decoding order. For example, the video decoder 30 may receive a bitstream from the video encoder 20 and buffer the AU and its corresponding DU in the CPB 94 for extraction at a determined removal time. For example, the video decoder 30 may decode a duration between CPB removal of a first DU in the AU from the CPB 94 and CPB removal of a second DU from the CPB 94, wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU.

The method further includes determining a removal time for the first DU based at least in part on the coded duration (202). For example, video decoder 30 may determine the CPB removal time for the first DU based on the decoded duration between the CPB removal of the first DU and the CPB removal of the second DU in the AU. Video decoder 30 may extract the DU from CPB 94 at approximately the determined CPB removal time.

In some examples, the method of FIG. 5 further includes decoding sub-picture-level CPB parameters, wherein determining the removal time of the first DU includes determining the removal time of the first DU based at least in part on the decoded duration and the sub-picture-level CPB parameters. Decoding the sub-picture-level CPB parameters may include decoding a sub-picture timing supplemental enhancement information (SEI) message associated with the first DU.

In an example where the second DU is the last DU in decoding order in the AU, the coded sub-picture SEI message includes the duration between the removal time of the last DU in the decoded sub-picture timing SEI message and the removal time of the first DU. In some examples, the sequence level flag is decoded to determine the presence of sub-picture level CPB parameters in the picture timing SEI message or in the sub-picture timing SEI message. For example, in response to decoding the sequence level flag and determining that the sub-picture level CPB parameters are present in the picture timing SEI message, video decoder 30 may decode the picture timing SEI message to parse the sub-picture level CPB parameters. Similarly, in response to decoding the sequence level flag and determining that the sub-picture level CPB parameters are present in the sub-picture timing SEI message, video decoder 30 may decode the sub-picture timing SEI message to parse the sub-picture level CPB parameters.

In some examples, determining the removal time of the first DU includes determining the removal time of the first DU without decoding an initial CPB removal delay and offset. When the AU has a TemporalId less than or equal to 0, the method may further include decoding at least one of a buffering period SEI message or a recovery point SEI message associated with the AU.

The decoding units described herein may be any decoding unit, as well as non-VCL network abstraction layer (NAL) units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63. Thus, video decoder 30 may decode a DU including a non-VCL NAL unit with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63 according to the techniques described in this disclosure.

In another example, the method includes deriving at least one of a CPB arrival time and a CPB nominal removal time for the AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether the first DU is an AU. For example, video decoder 30 derives the CPB arrival time and the CPB nominal removal time for the AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether the first DU is an AU.

In some examples, the method includes deriving at least one of a CPB arrival time and a CPB nominal removal time for an AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether a first DU is an AU. The method may include deriving the CPB removal time for the AU level when the syntax element indicates that the first DU is an AU. Deriving the CPB removal time for the AU level may include deriving the CPB removal time only for the AU level when the syntax element indicates that the DU is an AU.

The syntax element may be SubPicCpbFlag, wherein when SubPicCpbFlag is equal to 0, the DU is an AU, otherwise the DU includes one or more video coding layer (VCL) network abstraction layer (NAL) units and associated non-VCL NAL units in the AU. In some examples, the syntax element includes a first syntax element, and wherein deriving the CPB arrival time and the CPB nominal removal time includes deriving the CPB arrival time and the CPB nominal removal time when a second syntax element specifies that a sub-picture level CPB removal delay parameter is present and the CPB can operate at an access unit level or a sub-picture level.

The second syntax element may be sub_pic_cpb_params_present_flag, wherein when sub_pic_cpb_params_present_flag is equal to 1, the sub-picture level CPB removal delay parameter is present and the CPB can operate at the access unit level or the sub-picture level, and when sub_pic_cpb_params_present_flag is equal to 0, the sub-picture level CPB removal delay parameter is not present and the CPB operates at the access unit level. The second syntax element specifies that the sub-picture level CPB removal delay parameter is present and the CPB can operate at the AU level or the sub-picture level, and the method may further include determining a variable subPicParamsPresentFlag to be equal to 0, deriving an AU initial arrival time and an AU last arrival time, determining a variable subPicParamsPresentFlag to be equal to 1, and deriving a DU initial arrival time and a DU last arrival time for decoding units within the access unit.

When the syntax element indicates that the DU is not an AU, a CPB removal time for the sub-picture level may also be derived. Deriving the CPB removal time for the sub-picture level may include deriving the CPB removal time only for the sub-picture level when the syntax element indicates that the DU is not an AU. For example, when the syntax element indicates that the DU is not an AU, video decoder 30 may derive the CPB removal time for the sub-picture level.

FIG6 is a flowchart illustrating another method for determining a coded picture buffer (CPB) removal time for a first decoding unit in an access unit based on a CPB removal time for a second decoding unit of the access unit according to the techniques described in this disclosure. A video encoder may perform the method of FIG6. For example, the video encoder may be video encoder 20 of FIG1 or FIG2.

The method includes determining, for an AU that includes a first decoding unit (DU), a CPB removal time for a second DU, wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU (210). The second DU may immediately follow the first DU in the AU in decoding order. Alternatively, the second DU may be the last DU in the AU in decoding order. In some examples, video encoder 20 schedules the CPB removal time for the AU. In some examples, the CPB removal time is scheduled by a device external to video encoder 20 and the schedule is provided to video encoder 20.

The CPB removal time for an AU may be the same as the CPB removal time for the last DU in the AU. Thus, video encoder 20 may determine the CPB removal time for a second DU based on the scheduled CPB removal time for the AU. In some examples, determining the CPB removal time for the second DU based on the scheduled CPB removal time for the AU includes determining how many DUs are included in the AU and determining the scheduled time for each CPB. For example, according to the techniques described herein, video encoder 20 may determine the CPB removal time for a second DU for an AU that includes a first decoding unit DU, where the second DU follows the first DU in decoding order and is in the same AU as the first DU.

The method further includes determining a duration between a CPB removal time of the first DU and a determined CPB removal time of the second DU (212). For example, video encoder 20 may determine the duration between the CPB removal times of the first DU based on the scheduled CPB removal time of the AU and the number of DUs in the AU. In some examples, video encoder 20 determines the duration based on the scheduled CPB removal time for each DU in the AU.

The method further includes encoding the determined duration (214). Video encoder 20 may, for example, encode the determined duration as a syntax element in a sub-picture level CPB parameter set. For example, the method may further include encoding sub-picture level CPB parameters, wherein encoding the determined duration includes encoding the determined duration as one or more sub-picture level CPB parameters. Encoding the sub-picture level CPB parameters may include encoding a sub-picture timing supplemental enhancement information (SEI) message associated with the first DU. In one example, encoding the determined duration as the one or more sub-picture level CPB parameters further includes encoding the determined duration in the sub-picture timing SEI message.

In an example where the second DU is the last DU in decoding order in the AU, the encoded sub-picture SEI message includes a duration between a removal time of the last DU in the encoded sub-picture timing SEI message and a removal time of the first DU. In some examples, a sequence level flag is encoded to indicate the presence of sub-picture level CPB parameters in a picture timing SEI message or in a sub-picture timing SEI message. In some examples, determining the removal time of the first DU includes determining the removal time of the first DU without encoding an initial CPB removal delay and offset.

When the AU has a TemporalId less than or equal to 0, the method may further include encoding at least one of a buffering period SEI message or a recovery point SEI message associated with the AU.

The DUs described herein may be any type of DU, as well as non-VCL network abstraction layer (NAL) units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63. Video encoder 20 may encode any DU according to the techniques described in this disclosure, including DUs that are non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

In another example, the method includes deriving at least one of a CPB arrival time and a CPB nominal removal time for the AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether the first DU is an AU.

In some examples, the method includes deriving at least one of a CPB arrival time and a CPB nominal removal time for an AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether a first DU is an AU. The method may include deriving the CPB removal time for the AU level when the syntax element indicates that the first DU is an AU. Deriving the CPB removal time for the AU level may include deriving the CPB removal time only for the AU level when the syntax element indicates that the DU is an AU.

The syntax element may be SubPicCpbFlag, wherein when SubPicCpbFlag is equal to 0, the DU is an AU, otherwise the DU includes one or more video coding layer (VCL) network abstraction layer (NAL) units and associated non-VCL NAL units in the AU. In some examples, the syntax element includes a first syntax element, and wherein deriving the CPB arrival time and the CPB nominal removal time includes deriving the CPB arrival time and the CPB nominal removal time when a second syntax element specifies that a sub-picture level CPB removal delay parameter is present and the CPB may operate at an access unit level or a sub-picture level.

The second syntax element may be sub_pic_cpb_params_present_flag, wherein when sub_pic_cpb_params_present_flag is equal to 1, the sub-picture level CPB removal delay parameter is present and the CPB can operate at the access unit level or the sub-picture level, and when sub_pic_cpb_params_present_flag is equal to 0, the sub-picture level CPB removal delay parameter is not present and the CPB operates at the access unit level. The second syntax element specifies that the sub-picture level CPB removal delay parameter is present and the CPB can operate at the AU level or the sub-picture level, and the method may further include setting a variable subPicParamsPresentFlag to 0, deriving an AU initial arrival time and an AU last arrival time, setting a variable subPicParamsPresentFlag to 1, and deriving a DU initial arrival time and a DU last arrival time for decoding units within the access unit.

When the syntax element indicates that the DU is not an AU, a CPB removal time for the sub-picture level may also be derived. Deriving the CPB removal time for the sub-picture level may include deriving the CPB removal time only for the sub-picture level when the syntax element indicates that the DU is not an AU. For example, when the syntax element indicates that the DU is not an AU, video encoder 20 may derive the CPB removal time for the sub-picture level.

FIG7 is a flowchart illustrating a method for deriving a CPB removal time for a first DU based at least in part on a sub-picture timing SEI message according to the techniques described in this disclosure. The method may be performed by a video decoding device, for example, the video decoder 30 of FIG1 and FIG3.

The method includes decoding a sub-picture timing SEI message associated with a first decoding unit of an access unit (300). For example, according to the techniques described herein, video decoder 30 may decode a bitstream including coded data and corresponding syntax elements and a sub-picture timing SEI message associated with a first DU of an AU. Video decoder 30 may buffer the AU and its corresponding DU in CPB 94 for retrieval at a determined removal time. For example, video decoder 30 may decode the sub-picture timing SEI message associated with the first DU of the AU.

In some examples, the method includes decoding a sequence level flag to determine the presence of sub-picture level CPB parameters in a sub-picture timing SEI message or a picture timing SEI message associated with the first DU. The method may further include decoding the sub-picture level CPB parameters, wherein determining the CPB removal time for the first DU is further based at least in part on the sub-picture level CPB parameters. In response to receiving the encoded bitstream, video decoder 30 may decode the sequence level flag and determine from the value of the sequence level flag whether the sub-picture level CPB parameters are found in a sub-picture timing SEI message or a picture timing SEI message. Based on the value of the sequence level flag, video decoder 30 may decode the sub-picture timing SEI message or the picture timing SEI message to decode the sub-picture level CPB parameters.

In instances where the sequence-level flag indicates that sub-picture-level CPB parameters will be present in the sub-picture timing SEI message, decoding the sub-picture-level CPB parameters may include decoding the sub-picture timing SEI message associated with the first DU. In instances where the second DU is the last DU in decoding order in the AU, decoding the sub-picture SEI message may further include decoding the duration between the removal time of the last DU in the sub-picture timing SEI message and the removal time of the first DU.

The method further includes decoding a duration between coded picture buffer (CPB) removal of a second DU in decoding order and CPB removal of a first DU of the AU in a sub-picture SEI message, wherein the duration is in a sub-picture timing SEI message (302). For example, from the received bitstream, video decoder 30 may decode the duration between coded picture buffer (CPB) removal of a second DU in decoding order and CPB removal of a first DU of the AU in the sub-picture SEI message.

The method also includes deriving a CPB removal time for the first DU based at least in part on the sub-picture timing SEI message (304).Video decoder 30 may extract the first DU from CPB 94 for decoding at the determined CPB removal time.

In some examples, the second DU is the last DU in the AU in decoding order. The second DU may immediately follow the first DU in the AU in decoding order. In some examples, determining the removal time of the first DU includes determining the removal time of the first DU without decoding the initial CPB removal delay and offset.

A DU may be any type of decoding unit, including a non-video coding layer (VCL) network abstraction layer (NAL) unit with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

The techniques described herein may provide for more error-resilient determination of coded picture buffer removal time. Furthermore, in addition to improved error resilience, the techniques may also facilitate signaling efficiency, thereby reducing bandwidth, reducing signaling overhead, and increasing decoding time. Furthermore, the techniques described in this disclosure may allow for appropriate temporal scalability.

Such techniques may include, for example, determining a coded picture buffer removal time for a DU of an AU independently of the removal time of any other access units. For example, the CPB removal time for a DU of an AU may be signaled based on the duration between the CPB removal times of the next DU in decoding order in the AU or the duration between the CPB removal times of the last DU in the AU. According to the techniques described herein, the techniques may also include signaling a sequence-level flag to control the presence of sub-picture CPB parameters in only one of a picture timing SEI message or a sub-picture timing SEI message. The techniques may also include extending the definition of decoding units. Additional techniques provide for restricting buffering period SEI messages and recovery point SEI messages so that they cannot be associated with AUs with a variable TemporalId greater than 0. The techniques may also include providing a flag to signal whether CPB removal times are derived at the AU level or the sub-picture level.

FIG8 is a flowchart illustrating another method for deriving a CPB removal time for a first DU based at least in part on an encoded sub-picture timing SEI message according to the techniques described in this disclosure. The method may be performed by a video encoding device, for example, the video encoder 20 of FIG1 and FIG2.

The method includes determining a duration between a coded picture buffer (CPB) removal time of a first decoding unit (DU) in an access unit (AU) and a CPB removal time of a second DU in the AU (310). The duration may be determined, for example, based on subtracting a scheduled CPB removal time for the first DU from a scheduled CPB removal time for the second DU.

The method further includes encoding the duration in a sub-picture timing supplemental enhancement information (SEI) message associated with the AU (312). For example, video encoder 20 may encode the duration between coded picture buffer (CPB) removal of the second DU in decoding order and CPB removal of the first DU of the AU in the sub-picture SEI message in the bitstream. For example, according to the techniques described herein, video encoder 20 may encode a bitstream including the coded data and corresponding syntax elements, a sub-picture timing SEI message associated with the first DU of the AU.

In some examples, the method of FIG8 includes encoding a sequence-level flag to indicate the presence of sub-picture-level CPB parameters in a sub-picture timing SEI message or a picture timing SEI message associated with the first DU. The method may further include decoding the sub-picture-level CPB parameters, wherein determining the CPB removal time for the first DU is further based at least in part on the sub-picture-level CPB parameters. For example, video encoder 20 may encode a sequence-level flag in the bitstream to indicate the presence of sub-picture-level CPB parameters in a sub-picture timing SEI message or a picture timing SEI message associated with the first DU. Video encoder 20 may further encode the sub-picture-level CPB parameters in the bitstream.

In instances where the sequence-level flag indicates that sub-picture-level CPB parameters are to be present in the sub-picture timing SEI message, encoding the sub-picture-level CPB parameters may include encoding a sub-picture timing SEI message associated with the first DU. In instances where the second DU is the last DU in decoding order in the AU, encoding the sub-picture SEI message may further include encoding, in the sub-picture timing SEI message, a duration between a removal time of the last DU and a removal time of the first DU.

In some examples, the second DU is the last DU in the AU in decoding order. The second DU may immediately follow the first DU in the AU in decoding order. In some examples, determining the removal time of the first DU includes determining the removal time of the first DU without encoding the initial CPB removal delay and offset.

A DU may be any type of decoding unit, including a non-video coding layer (VCL) network abstraction layer (NAL) unit with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

FIG9 is a flowchart illustrating a method for decoding a sequence level flag for sub-picture level coded picture buffer parameters according to the techniques described in this disclosure. The method may be performed by a video decoding device, for example, the video decoder 30 of FIG1 and FIG3.

The method includes decoding a sequence level flag to determine the presence of one or more sub-picture level CPB parameters for a DU of an AU in a picture timing SEI message or a sub-picture timing SEI message associated with the DU (400). For example, video decoder 30 decodes the sequence level flag to determine the presence of the one or more sub-picture level CPB parameters. Video decoder 30 also decodes the sequence level flag to determine the location of the one or more sub-picture level CPB parameters. The sequence level flag may be sub_pic_cpb_params_present_flag. In some examples, the one or more sub-picture level CPB parameters are present in only one of the picture timing SEI message or the sub-picture timing SEI message.

The method may further include decoding one or more sub-picture level CPB parameters from a picture timing SEI message or a sub-picture timing SEI message based on the sequence level flag (402). For example, in response to the sequence level flag indicating that the one or more sub-picture level CPB parameters are present in the picture timing SEI message, video decoder 30 decodes the picture timing SEI message to determine the one or more sub-picture level CPB parameters. Similarly, in response to the sequence level flag indicating that the one or more sub-picture level CPB parameters are present in the sub-picture timing SEI message, video decoder 30 decodes the sub-picture timing SEI message to determine the one or more sub-picture level CPB parameters.

The method may further include determining a CPB removal time for the DU based at least in part on the one or more sub-picture level CPB parameters. In some examples, determining the CPB removal time for the DU includes determining the CPB removal time for the DU without decoding an initial CPB removal delay and offset.

In an example where the sequence-level flag indicates that the sub-picture-level CPB parameters are present in the sub-picture timing SEI message, decoding the sub-picture-level CPB parameters may include decoding the sub-picture timing SEI message associated with the DU. In another example, the method may include deriving at least one of a CPB arrival time and a CPB nominal removal time for the AU at both the access unit level and the sub-picture level, regardless of the value of the syntax element defining whether the first DU is an AU. That is, video decoder 30 may derive at least one of a CPB arrival time and a CPB nominal removal time for the AU at both the access unit level and the sub-picture level.

In another example, the DU is a first DU, and the method further includes deriving a CPB removal time for the first DU based at least in part on sub-picture-level CPB parameters and a duration between the CPB removal time of a second DU in decoding order of the decoded AU and the CPB removal time of the first DU. The method may further include decoding video data of the first DU based at least in part on the CPB removal time. In some examples, the second DU is the last DU in decoding order in the AU, or immediately follows the first DU in decoding order.

DU can be any DU, including non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC12.

In examples where the AU has a TemporalId value equal to 0, the method may further include decoding at least one of a buffering period SEI message or a recovery point SEI message associated with the AU. For example, video decoder 30 may decode at least one of a buffering period SEI message or a recovery point SEI message associated with the AU having a TemporalId value equal to 0.

FIG10 is a flowchart illustrating a method for encoding a sequence level flag for sub-picture level coded picture buffer parameters according to the techniques described in this disclosure. The method may be performed by a video encoding device, for example, the video encoder 20 of FIG1 and FIG2.

The method includes encoding one or more sub-picture level coded picture buffer (CPB) parameters for a decoding unit (DU) of an access unit (AU) in a picture timing SEI message or a sub-picture timing SEI message (410). Video encoder 20 may encode the one or more sub-picture level CPB parameters in the picture timing SEI message. Alternatively, video encoder 20 may encode the one or more sub-picture level CPB parameters in the sub-picture timing SEI message.

The method further includes encoding a sequence level flag to indicate the presence of one or more sub-picture level CPB parameters for a DU of an AU in a picture timing SEI message or a sub-picture timing SEI message associated with the DU (412). For example, video encoder 20 encodes a sequence level flag to indicate the presence and location of the one or more sub-picture level CPB parameters. The sequence level flag may be sub_pic_cpb_params_present_flag. In some examples, video encoder 20 encodes the one or more sub-picture level CPB parameters in only one of the picture timing SEI message or the sub-picture timing SEI message.

The method may further include determining a CPB removal time for the DU based at least in part on the one or more sub-picture level CPB parameters. In some examples, determining the CPB removal time for the DU includes determining the CPB removal time for the DU without encoding an initial CPB removal delay and offset.

In one example, encoding the one or more sub-picture level CPB parameters further includes encoding the one or more sub-picture level CPB parameters in a sub-picture timing SEI message associated with the DU. In this example, video encoder 20 encodes a sequence level flag to indicate that the sub-picture level CPB parameters are present in the sub-picture timing SEI message. In another example, encoding the one or more sub-picture level CPB parameters further includes encoding the one or more sub-picture level CPB parameters in a picture timing SEI message associated with the DU. In this example, video encoder 20 encodes a sequence level flag to indicate that the sub-picture level CPB parameters are present in the picture timing SEI message.

In another example, the DU is a first DU, and the method further includes deriving a CPB removal time for the first DU based at least in part on sub-picture-level CPB parameters, and encoding a duration between the CPB removal time of a second DU in decoding order of the AU and the CPB removal time of the first DU. The method may further include encoding video data of the first DU based at least in part on the CPB removal time. In some examples, the second DU is the last DU in decoding order in the AU, or immediately follows the first DU in decoding order.

DU can be any DU, including non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC12.

In examples where the AU has a TemporalId value equal to 0, the method may further include encoding at least one of a buffering period SEI message or a recovery point SEI message associated with the AU. For example, video encoder 20 may encode at least one of a buffering period SEI message or a recovery point SEI message associated with the AU having a TemporalId value equal to 0.

FIG11 is a flowchart illustrating a method for decoding a DU with an extended definition according to the techniques described in this disclosure. The method may be performed by a video decoding device, for example, the video decoder 30 of FIG1 and FIG3.

The method includes decoding a duration between coded picture buffer (CPB) removal of a first decoding unit (DU) in an access unit (AU) and CPB removal of a second DU, wherein the first DU comprises a non-video coding layer (VCL) network abstraction layer (NAL) unit with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63 (500). That is, in addition to other DU types defined in HEVC WD8, video decoder 30 can also decode DUs with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63 as non-video coding layer (VCL) network abstraction layer (NAL) units.

In some examples, the second DU follows the first DU in decoding order and is in the same AU as the first DU. The second DU may immediately follow the first DU in the AU in decoding order. In other examples, the second DU is the last DU in the AU in decoding order.

The method also includes determining a removal time for the first DU based at least in part on the decoded duration (502). The method further includes decoding video data of the first DU based at least in part on the removal time (504). For example, video decoder 30 determines the removal time for the first DU based in part on the decoded duration and then decodes the video data of the first DU based on the removal time.

In one example, the method may further include decoding one or more sub-picture level CPB parameters, wherein determining the removal time of the first DU comprises determining the removal time of the first DU based at least in part on the decoded duration and the sub-picture level CPB parameters. Decoding the one or more sub-picture level CPB parameters may further include decoding a sub-picture timing supplemental enhancement information (SEI) message associated with the first DU.

In another example where the second DU is the last DU in decoding order in the AU, the decoded sub-picture SEI message includes the duration between the removal time of the last DU in the decoded sub-picture timing SEI message and the removal time of the first DU. In some examples, video decoder 30 decodes the sequence level flag to determine the presence of sub-picture level CPB parameters in the picture timing SEI message or in the sub-picture timing SEI message.

In another example where the AU has a TemporalId equal to 0, video decoder 30 may decode at least one of a buffering period SEI message or a recovery point SEI message associated with the AU. The method may also include deriving at least one of a CPB arrival time and a CPB nominal removal time for the AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether the first DU is an AU.

FIG12 is a flowchart illustrating a method for encoding a DU with an extended definition according to the techniques described in this disclosure. The method may be performed by a video encoding device, for example, the video encoder 20 of FIG1 and FIG2.

The method includes determining a CPB removal time for a second DU for an AU including a first DU, wherein the second DU follows the first DU in decoding order and is in the same AU as the first DU, and wherein the first DU includes a non-video coding layer (VCL) network abstraction layer (NAL) unit (510) with a nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63. That is, in addition to other DU types defined in HEVC WD8, video encoder 20 may also encode a DU with a nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63 as a non-VCL NAL unit. The second DU may follow (including immediately follow) the first DU in decoding order and be in the same AU as the first DU. In other examples, the second DU is the last DU in the AU in decoding order.

The method also includes determining a duration between the CPB removal time of the first DU and the determined CPB removal time of the second DU (512). The determination of the duration between the CPB removal time of the first DU and the determined CPB removal time of the second DU may be based on the scheduled CPB removal time of the AU. The method further includes encoding the determined duration (514). For example, video encoder 20 determines the duration between the CPB removal of the first DU and the second DU, and then encodes the determined duration as a syntax element.

In one example, the method may further include encoding one or more sub-picture level CPB parameters, wherein determining the determined duration of the first DU includes determining a removal time of the first DU based at least in part on the decoded duration and the sub-picture level CPB parameters. Encoding the one or more sub-picture level CPB parameters may further include encoding a sub-picture timing SEI message associated with the first DU.

In another example where the second DU is the last DU in decoding order in the AU, encoding the sub-picture SEI message includes encoding the duration between the removal time of the last DU and the removal time of the first DU in the sub-picture timing SEI message. In some examples, video encoder 20 encodes a sequence level flag to indicate the presence of sub-picture level CPB parameters in a picture timing SEI message or in a sub-picture timing SEI message.

In another example where the AU has a TemporalId equal to 0, video encoder 20 may encode at least one of a buffering period SEI message or a recovery point SEI message associated with the AU. The method may also include deriving at least one of a CPB arrival time and a CPB nominal removal time for the AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether the first DU is an AU.

FIG13 is a flowchart illustrating a method for decoding a buffering period and recovery point SEI message according to the techniques described in this disclosure. The method may be performed by a video decoding device, for example, the video decoder 30 of FIG1 and FIG3.

The method includes decoding a buffering period SEI message associated with an AU (530). The AU has a temporalId equal to or less than 0. That is, the buffering period SEI message is restricted such that it cannot be associated with an AU having a temporalId greater than 0.

The method further includes decoding, from the buffering period SEI message, a duration between removal of the CPB of a first DU and removal of the CPB of a second DU in the AU (532). The second DU may be in the same AU as the first DU. The second DU may be after (including immediately after) the first DU in decoding order. In other examples, the second DU may be the last DU in decoding order in the AU. The DU may be any DU type accepted in HEVC WD8 and may further be a VCL NAL unit with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

The method may further include determining a removal time of the first DU based at least in part on the decoded duration (534). In some examples, video decoder 30 may decode one or more sub-picture level CPB parameters. Determining the removal time of the first DU may further include determining the removal time of the first DU based at least in part on the decoded duration and the sub-picture level CPB parameters. Decoding the one or more sub-picture level CPB parameters may further include decoding a sub-picture timing SEI message associated with the first DU.

The method may further include decoding video data of the first DU based at least in part on the removal time (536). In an example where the second DU is the last DU in decoding order in the AU, the decoded sub-picture SEI message further includes a duration between the removal time of the last DU in the decoded sub-picture timing SEI message and the removal time of the first DU.

The method may further include decoding a sequence level flag to determine the presence of sub-picture level CPB parameters in a picture timing SEI message or in a sub-picture timing SEI message. The method may also include deriving at least one of a CPB arrival time and a CPB nominal removal time for an AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether the first DU is an AU.

FIG14 is a flowchart illustrating a method for encoding a buffering period SEI message according to the techniques described in this disclosure. The method may be performed by a video encoding device, for example, the video encoder 20 of FIG1 and FIG2.

The method includes encoding a buffering period supplemental enhancement information (SEI) message associated with an access unit (AU), wherein the duration is encoded within at least one of a buffering period SEI message or a recovery point SEI message (540). Because the AU has a temporalId equal to or less than 0, the buffering period SEI message is restricted such that it cannot be associated with an AU having a temporalId greater than 0.

The method may also include encoding a duration between a CPB removal time of a first DU and a CPB removal time of a second DU in an AU from a buffering period SEI message, wherein the AU has a temporalId equal to 0 (542). The method may further include determining a removal time of the first DU based at least in part on the decoded duration (544). Additionally, the method may include encoding video data of the first DU (546).

The method may further include determining a duration between removal of a coded picture buffer (CPB) of a first decoding unit (DU) in an access unit (AU) and removal of the CPB of a second DU in the AU, wherein the AU has a TemporalId equal to 0. The second DU may be in the same AU as the first DU. The second DU may be after (including immediately after) the first DU in decoding order. In other examples, the second DU may be the last DU in the AU in decoding order. The DU may be any DU type accepted in HEVC WD8 and may further be a VCL NAL unit with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

In one example, determining the duration between CPB removal of the first DU may include determining removal times of the first and second DUs. The removal time of the first DU may be subtracted from the removal time of the second DU to determine the duration.

In some examples, video encoder 20 may encode one or more sub-picture level CPB parameters. Determining the removal time of the first DU may further include determining the removal time of the first DU based at least in part on the encoded duration and the sub-picture level CPB parameters. Encoding the one or more sub-picture level CPB parameters may further include encoding a sub-picture timing SEI message associated with the first DU.

The method may further include encoding video data of the first DU. Encoding the video data of the first DU may be based at least in part on the removal time. In an example where the second DU is the last DU in decoding order in the AU, encoding the sub-picture SEI message further includes encoding, in the sub-picture timing SEI message, a duration between the removal time of the last DU and the removal time of the first DU.

The method may further include encoding a sequence level flag to indicate the presence of sub-picture level CPB parameters in a picture timing SEI message or in a sub-picture timing SEI message. The method may also include deriving at least one of a CPB arrival time and a CPB nominal removal time for an AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element defining whether the first DU is an AU.

15 is a flowchart illustrating a method for decoding coded picture buffer arrival and nominal removal times according to the techniques described in this disclosure. The method may be performed by a video decoding device, for example, the video decoder 30 of FIGS. 1 and 3 .

The method includes deriving at least one of a CPB arrival time and a CPB nominal removal time for an AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element that defines whether the DU is an entire AU. The DU may be associated with the AU (560). The method may include video decoder 30 determining a value of a syntax element. The syntax element may have the form of a SubPicCpbFlag. In response to the syntax element having a true value (e.g., SubPicCpbFlag is 1), the method may include deriving the CPB removal time only for the AU level. In response to the syntax element having a false value (e.g., SubPicCpbFlag is 0), the CPB removal time is derived only for the sub-picture level. In some examples, at least one of the CPB arrival time and the CPB nominal removal time is derived only when a syntax flag indicating the presence of CPB parameters has a true value.

The method may further include determining a removal time of the AU based at least in part on one of a CPB arrival time and a CPB nominal removal time (562).The method further includes decoding video data of the AU based at least in part on the removal time (564).

The method may further include decoding a duration between CPB removal of a first DU and CPB removal of a second DU in the AU, determining a removal time of the first DU based at least in part on the decoded duration, and decoding video data of the first DU based at least in part on at least one of the removal time, a CPB arrival time, and a CPB nominal removal time. In some examples, the second DU follows the first DU in decoding order and is in the same AU as the first DU. The method may further include decoding one or more sub-picture-level CPB parameters, wherein determining the removal time of the first DU includes determining the removal time of the first DU based at least in part on the decoded duration and the sub-picture-level CPB parameters.

In some examples, the method also includes decoding a sequence level flag to determine presence of sub-picture level CPB parameters in a picture timing SEI message or in a sub-picture timing SEI message.

The DU may be any type of DU described in HEVC WD8, including non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

In another example where the AU has a TemporalId not greater than 0, the method further includes decoding at least one of a buffering period supplemental enhancement information (SEI) message or a recovery point SEI message associated with the AU.

16 is a flowchart illustrating a method for encoding coded picture buffer arrival and nominal removal times according to the techniques described in this disclosure. The method may be performed by a video encoding device, for example, the video encoder 20 of FIG. 1 and FIG. 2 .

The method includes deriving at least one of a CPB arrival time and a CPB nominal removal time for an AU at both the access unit level and the sub-picture level, regardless of a value of a syntax element that defines whether the DU is an entire AU. The DU may be associated with the AU (570). The method may include video encoder 20 determining a value of a syntax element. The syntax element may be in the form of SubPicCpbFlag. In response to the syntax element having a true value (e.g., SubPicCpbFlag is 1), video encoder 20 may derive the CPB removal time only for the AU level. In response to the syntax element having a false value (e.g., SubPicCpbFlag is 0), video encoder 20 may derive the CPB removal time only for the sub-picture level. In some examples, at least one of the CPB arrival time and the CPB nominal removal time is derived only when a syntax flag indicating the presence of CPB parameters has a true value.

The method may further include determining a removal time for the AU based at least in part on one of a CPB arrival time and a CPB nominal removal time (572). The method further includes encoding the determined removal time (574). In some examples, encoding the removal time may include encoding a duration between CPB removal of a first DU and CPB removal of a second DU in the AU, determining the removal time of the first DU based at least in part on the encoded duration, and encoding video data of the first DU based at least in part on at least one of the removal time, the CPB arrival time, and the CPB nominal removal time. In some examples, the second DU follows the first DU in decoding order and is in the same AU as the first DU. The method may further include encoding one or more sub-picture level CPB parameters, wherein determining the removal time for the first DU includes determining the removal time for the first DU based at least in part on the encoded duration and the sub-picture level CPB parameters. The method may further include encoding a duration between CPB removal of the first DU and CPB removal of the second DU in the AU, wherein encoding the removal time further includes the encoded duration.

In some examples, the method also includes encoding a sequence level flag to indicate presence of sub-picture level CPB parameters in a picture timing SEI message or in a sub-picture timing SEI message.

The DU may be any type of DU described in HEVC WD8, including non-VCL NAL units with nal_unit_type equal to UNSPEC0, EOS_NUT, EOB_NUT, in the range of RSV_NVCL44 to RSV_NVCL47, or in the range of UNSPEC48 to UNSPEC63.

In another example where the AU has a TemporalId not greater than 0, the method further includes encoding at least one of a buffering period SEI message or a recovery point SEI message associated with the AU.

In one or more examples, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on or transmitted via a computer-readable medium as one or more instructions or codes and executed by a hardware-based processing unit. Computer-readable media may include computer-readable storage media (which corresponds to tangible media such as data storage media) or communication media, which includes, for example, any media that facilitates the transfer of a computer program from one place to another according to a communication protocol. In this manner, computer-readable media may generally correspond to (1) non-transitory, tangible computer-readable storage media, or (2) communication media such as signals or carrier waves. Data storage media may be any available media that can be accessed by one or more computers or one or more processors to retrieve instructions, codes, and/or data structures for implementing the techniques described in this disclosure. A computer program product may include computer-readable media.

By way of example, and not limitation, such computer-readable storage media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, flash memory, or any other medium that can be used to store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly referred to as a computer-readable medium. For example, if instructions are transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies (such as infrared, radio, and microwave), then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies (such as infrared, radio, and microwave) are included in the definition of medium. However, it should be understood that computer-readable storage media and data storage media do not include connections, carrier waves, signals, or other transient media, but rather refer to non-transitory, tangible storage media. As used herein, disk and disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, and Blu-ray disc, where disks typically reproduce data magnetically, while discs reproduce data optically using lasers. Combinations of the above should also be included within the scope of computer-readable media.

Instructions may be executed by one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), field-programmable logic arrays (FPGAs), or other equivalent integrated or discrete logic circuits. Thus, the term "processor," as used herein, may refer to any of the foregoing structures or any other structure suitable for implementing the techniques described herein. Furthermore, in some aspects, the functionality described herein may be provided within dedicated hardware and/or software modules configured for encoding and decoding, or incorporated into a combined codec. Likewise, the techniques may be fully implemented in one or more circuits or logic elements.

The techniques of this disclosure can be implemented in a wide variety of devices or apparatuses, including wireless handsets, integrated circuits (ICs), or collections of ICs (e.g., chipsets). Various components, modules, or units are described in this disclosure to emphasize functional aspects of devices configured to perform the disclosed techniques, but do not necessarily require implementation by different hardware units. Instead, as described above, the various units may be combined in a codec hardware unit or provided by a collection of interoperable hardware units (including one or more processors as described above) in conjunction with appropriate software and/or firmware.

Various examples have been described. These and other examples are within the scope of the following claims.

Claims (36)

1. A method for decoding video data, the method comprising: A Sub-Picture Timing Supplement Enhancement (SEI) message is decoded. This SEI message provides decoded picture buffer (CPB) removal delay information for the decoding unit (DU) associated with the SEI message. The DU associated with the SEI message consists of the following units in decoding order: an SEI Network Abstraction Layer (NAL) unit containing the SEI message, and one or more NAL units that do not contain the SEI message. The DU associated with the SEI message includes all subsequent NAL units in the access unit (AU) up to any subsequent SEI NAL unit containing the additional SEI message, but the DU associated with the SEI message does not include any subsequent SEI NAL units containing the additional SEI message. The duration between the CPB removal time of the last DU in the decoding order of the AU and the CPB removal time of the DU associated with the sub-image SEI message is determined from the sub-image timing SEI message; The CPB removal time of the DU associated with the sub-image temporal SEI message is derived at least in part based on the determined duration; and At the CPB removal time of the DU associated with the sub-image time-series SEI message, decode the video data associated with the DU associated with the sub-image time-series SEI message. 2. The method according to claim 1, further comprising: Decode sequence level flags to determine the presence of one or more sub-picture level CPB parameters in the sub-picture time-series SEI message or picture time-series SEI message associated with the DU associated with the sub-picture time-series SEI message. 3. The method according to claim 2, further comprising: Decode the sub-image level CPB parameters, wherein the determination of the CPB removal time of the DU associated with the sub-image temporal SEI message is further based at least in part on the sub-image level CPB parameters. 4. The method according to claim 3, wherein the sequence level flag indicates that the sub-image level CPB parameters will exist in the sub-image temporal SEI message, and wherein decoding the sub-image level CPB parameters includes: Decode the sub-image timing SEI message associated with the DU that is associated with the sub-image timing SEI message. 5. The method of claim 1, wherein determining the removal time of the DU associated with the sub-image timing SEI message comprises determining the removal time of the DU associated with the sub-image timing SEI message without decoding the initial CPB removal delay and offset. 6. A video decoding apparatus comprising a decoded image buffer and a video decoder, the video decoder including one or more processors configured to perform the following operations: A Sub-Picture Timing Supplement Enhancement (SEI) message is decoded. This SEI message provides decoded picture buffer (CPB) removal delay information for the decoding unit (DU) associated with the SEI message. The DU associated with the SEI message consists of the following units in decoding order: an SEI Network Abstraction Layer (NAL) unit containing the SEI message, and one or more NAL units that do not contain the SEI message. The DU associated with the SEI message includes all subsequent NAL units in the access unit (AU) up to any subsequent SEI NAL unit containing the additional SEI message, but the DU associated with the SEI message does not include any subsequent SEI NAL units containing the additional SEI message. The duration between the CPB removal time of the last DU in the decoding order of the AU and the CPB removal time of the DU associated with the sub-image SEI message is determined from the sub-image timing SEI message; The CPB removal time of the DU associated with the sub-image temporal SEI message is derived at least in part based on the determined duration; and At the CPB removal time of the DU associated with the sub-image time-series SEI message, decode the video data associated with the DU associated with the sub-image time-series SEI message. 7. The video decoding apparatus of claim 6, wherein the video decoder is further configured to decode sequence level flags to determine the presence of one or more sub-picture level CPB parameters in the sub-picture time-series SEI message or picture time-series SEI message associated with the DU associated with the sub-picture time-series SEI message. 8. The video decoding apparatus of claim 7, wherein the video decoder is further configured to decode the sub-picture level CPB parameters, wherein the video decoder is configured to determine the CPB removal time of the DU associated with the sub-picture timing SEI message further based at least in part on the determined duration and the sub-picture level CPB parameters. 9. The video decoding apparatus of claim 8, wherein the sequence level flag indicates that the sub-picture level CPB parameter will exist in the sub-picture temporal SEI message, and wherein the video decoder is configured such that the video decoder decodes the sub-picture temporal SEI message associated with the DU associated with the sub-picture temporal SEI message as part of decoding the sub-picture level CPB parameter. 10. The video decoding apparatus of claim 6, wherein the video decoder is configured such that it determines the removal time of the DU associated with the sub-picture timing SEI message without decoding the initial CPB removal delay and offset. 11. A non-transitory computer-readable storage medium having instructions stored thereon, which, when executed, are used by a processor of an apparatus for decoding video data: A Sub-Picture Timing Supplement Enhancement (SEI) message is decoded. This SEI message provides decoded picture buffer (CPB) removal delay information for the decoding unit (DU) associated with the SEI message. The DU associated with the SEI message consists of the following units in decoding order: an SEI Network Abstraction Layer (NAL) unit containing the SEI message, and one or more NAL units that do not contain the SEI message. The DU associated with the SEI message includes all subsequent NAL units in the access unit (AU) up to any subsequent SEI NAL unit containing the additional SEI message, but the DU associated with the SEI message does not include any subsequent SEI NAL units containing the additional SEI message. The duration between the CPB removal time of the last DU in the decoding order of the AU and the CPB removal time of the DU associated with the sub-image SEI message is determined from the sub-image timing SEI message; The CPB removal time of the DU associated with the sub-image temporal SEI message is derived at least in part based on the determined duration; and At the CPB removal time of the DU associated with the sub-image time-series SEI message, decode the video data associated with the DU associated with the sub-image time-series SEI message. 12. The computer-readable storage medium of claim 11, wherein the instructions, when executed, further cause the processor to: Decode sequence level flags to determine the presence of one or more sub-picture level CPB parameters in the sub-picture time-series SEI message or picture time-series SEI message associated with the DU associated with the sub-picture time-series SEI message. 13. The computer-readable storage medium of claim 12, wherein the instructions, when executed, further cause the processor to: Decode the sub-image level CPB parameters, wherein the instructions, when executed, cause the processor to determine the CPB removal time of the DU associated with the sub-image timing SEI message, further at least in part based on the sub-image level CPB parameters. 14. The computer-readable storage medium of claim 13, wherein the sequence level flag indicates that the sub-picture level CPB parameter will be present in the sub-picture timing SEI message, and wherein, as part of causing the processor to decode the sub-picture level CPB parameter, the instruction, when executed, causes the processor to decode the sub-picture timing SEI message associated with the DU associated with the sub-picture timing SEI message. 15. A video decoding apparatus, comprising: A means for decoding a Sub-Picture Timing Supplement Enhancement (SEI) message, the sub-picture timing SEI message providing decoded picture buffer (CPB) removal delay information for a decoding unit (DU) associated with the sub-picture timing SEI message, the DU associated with the sub-picture timing SEI message comprising, in decoding order: an SEI Network Abstraction Layer (NAL) unit containing the sub-picture timing SEI message, and one or more subsequent NAL units not containing the sub-picture timing SEI message, wherein the DU associated with the sub-picture timing SEI message includes all subsequent NAL units in an access unit (AU) up to any subsequent SEI NAL unit containing the additional sub-picture timing SEI message, but the DU associated with the sub-picture timing SEI message does not include any subsequent SEI NAL units containing the additional sub-picture timing SEI message; A means for determining, from the sub-image timing SEI message, the duration between the CPB removal time of the last DU in the decoding order of the AU and the CPB removal time of the DU associated with the sub-image SEI message; A means for deriving the CPB removal time of the DU associated with the sub-image temporal SEI message, at least in part, based on the determined duration; and A means for decoding the video data associated with the DU associated with the sub-picture time-series SEI message at the CPB removal time of the DU associated with the sub-picture time-series SEI message. 16. The video decoding apparatus according to claim 15, further comprising: A means for decoding sequence level flags to determine the presence of one or more sub-picture level CPB parameters in the sub-picture time-series SEI message or picture time-series SEI message associated with the DU associated with the sub-picture time-series SEI message. 17. The video decoding apparatus according to claim 16, further comprising: A means for decoding the sub-image level CPB parameters, wherein determining the CPB removal time of the DU associated with the sub-image timing SEI message is further based at least in part on the sub-image level CPB parameters. 18. The video decoding apparatus of claim 17, wherein the sequence level flag indicates that the sub-picture level CPB parameter will exist in the sub-picture timing SEI message, and wherein the means for decoding the sub-picture level CPB parameter includes means for decoding the sub-picture timing SEI message associated with the DU associated with the sub-picture timing SEI message. 19. A method for encoding video data, the method comprising: Determine the removal time of the decoded image buffer CPB of the first decoding unit DU in the access unit AU; Determine the duration between the CPB removal time of the first DU in the AU and the CPB removal time of the second DU in the AU; and Encoding the first DU includes encoding the duration in a sub-picture timing supplemental enhancement information (SEI) message associated with the AU, wherein the first DU in the AU and in decoding order consists of the following units: an SEI network abstraction layer (NAL) unit containing the sub-picture timing SEI message in which the duration is encoded, and one or more NAL units thereafter that do not contain the sub-picture timing SEI message, and wherein the first DU includes all subsequent NAL units in the AU up to any subsequent SEI NAL unit containing the additional sub-picture timing SEI message, but the first DU does not include any subsequent SEI NAL units containing the additional sub-picture timing SEI message. 20. The method of claim 19, further comprising encoding one or more sub-image level CPB parameters. 21. The method of claim 20, further comprising: Encoded sequence level flags are used to indicate the presence of one or more sub-picture level CPB parameters in the sub-picture timing SEI message or picture timing SEI message associated with the first DU. 22. The method of claim 21, wherein the sequence level flag indicates that the sub-image level CPB parameter exists in the sub-image temporal SEI message, and wherein encoding the one or more sub-image level CPB parameters includes encoding the sub-image temporal SEI message associated with the AU. 23. The method of claim 19, wherein the second DU is the last DU in the AU in the decoding order. 24. The method of claim 19, wherein the second DU follows immediately after the first DU in the AU in the decoding order. 25. The method of claim 19, wherein determining the duration comprises determining the removal time of the first DU without encoding the initial CPB removal delay and offset. 26. A video encoding apparatus comprising a decoded image buffer (CPB) and a video encoder configured to: Determine the CPB removal time of the first decoding unit DU in the access unit AU; Determine the duration between the CPB removal time of the first DU in the AU and the CPB removal time of the second DU in the AU; and Encoding the first DU includes encoding the duration in a sub-picture timing supplemental enhancement information (SEI) message associated with the AU, wherein the first DU in the AU and in decoding order consists of the following units: an SEI network abstraction layer (NAL) unit containing the sub-picture timing SEI message in which the duration is encoded, and one or more NAL units thereafter that do not contain the sub-picture timing SEI message, and wherein the first DU includes all subsequent NAL units in the AU up to any subsequent SEI NAL unit containing the additional sub-picture timing SEI message, but the first DU does not include any subsequent SEI NAL units containing the additional sub-picture timing SEI message. 27. The video encoding apparatus of claim 26, wherein the video encoder is further configured to encode one or more sub-picture level CPB parameters. 28. The video encoding apparatus of claim 27, wherein the video encoder is further configured to encode sequence level flags to indicate the presence of the one or more sub-picture level CPB parameters in the sub-picture timing SEI message or picture timing SEI message associated with the first DU. 29. The video encoding apparatus of claim 28, wherein the sequence level flag indicates that the sub-picture level CPB parameter exists in the sub-picture temporal SEI message, and wherein the video encoder is configured to encode the sub-picture temporal SEI message associated with the AU. 30. The video encoding apparatus of claim 26, wherein the second DU is the last DU in the AU in the decoding order. 31. The video encoding apparatus of claim 26, wherein the second DU follows immediately after the first DU in the AU in the decoding order. 32. The video encoding apparatus of claim 26, wherein the video encoder is configured to determine the removal time of the first DU without encoding the initial CPB removal delay and offset. 33. A non-transitory computer-readable storage medium having instructions stored thereon, which, when executed, are used by a processor of a device for encoding video data: Determine the removal time of the decoded image buffer CPB of the first decoding unit DU in the access unit AU; Determine the duration between the CPB removal time of the first DU in the AU and the CPB removal time of the second DU in the AU; and Encoding the first DU includes encoding the duration in a sub-picture timing supplemental enhancement information (SEI) message associated with the AU, wherein the first DU in the AU and in decoding order consists of the following units: an SEI network abstraction layer (NAL) unit containing the sub-picture timing SEI message in which the duration is encoded, and one or more NAL units thereafter that do not contain the sub-picture timing SEI message, and wherein the first DU includes all subsequent NAL units in the AU up to any subsequent SEI NAL unit containing the additional sub-picture timing SEI message, but the first DU does not include any subsequent SEI NAL units containing the additional sub-picture timing SEI message. 34. The computer-readable storage medium of claim 33, wherein the instructions, when executed, further cause the processor to: Encode one or more sub-image layer CPB parameters. 35. The computer-readable storage medium of claim 34, wherein the instructions, when executed, further cause the processor to: Encoded sequence level flags are used to indicate the presence of one or more sub-picture level CPB parameters in the sub-picture timing SEI message or picture timing SEI message associated with the first DU. 36. The computer-readable storage medium of claim 35, wherein the sequence level flag indicates that the sub-picture level CPB parameters are present in the sub-picture timing SEI message, and wherein, as part of causing the processor to encode the one or more sub-picture level CPB parameters, the instruction causes the processor to encode the sub-picture timing SEI message associated with the AU.