WO2016011823A1 - Method for acquiring live video slice, server, and storage medium - Google Patents

Abstract

Disclosed is a method for acquiring a live video slice. The method comprises: acquiring a reference time length of a slice; determining a sum of the reference time length and a first preset time length as a first time length, and/or determining a difference between the reference time length and the first preset time length as a second time length; acquiring a q^th key frame of a current original live source; determining a start position of the q^th key frame as a start position of a q^th slice; in a first time length and/or a second time length of the start position of the q^th key frame, when it is determined that a (q+1)^th key frame exists, determining a start position of the (q+1)^th key frame as an end position of the q^th slice; and at the start position and an end position of the q^th frame, slicing the current original live source to form the q^th slice, q being a positive integer. Also disclosed are a server and a computer storage medium.

Description

Method, server and storage medium for obtaining live video slice Technical field

The present invention relates to a live video streaming media technology, and in particular, to a method, a server and a computer storage medium for acquiring a live video slice.

Background technique

The HLS (HTTP Live Streaming) protocol is a dynamic code stream adaptation protocol based on the Hypertext Transfer Protocol (HTTP). It is a streaming media transmission protocol proposed by Apple. In recent years, the HLS protocol has been widely used in the field of streaming media, mainly because it uses HTTP protocol transmission, and can adaptively adjust the code rate according to the network bandwidth, such as allowing the server on the network side to store multiple different code rates for the same program content. The multimedia stream, and the terminal can adaptively adjust the code rate of the multimedia stream to be downloaded according to the actual network download speed; when the network bandwidth is tight, the terminal can achieve uninterrupted playback by reducing the code rate.

Compared with other common live streaming protocols, such as the Routing Table Maintenance Protocol (RTMP), the Real Time Streaming Protocol (RTSP), and the Microsoft Media Server Protocol (MMS). Etc., using the HLS protocol to broadcast live video, the biggest difference is that the multimedia data acquired by the terminal is not a complete data stream, but a slice with data recorded, because: in the HLS protocol, the server side The live data stream is stored as a continuous, short-lived media file (Motion Picture Experts Group MPEG-Transport Stream TS format). These media files are called slices, and the client only needs to continuously download and play these slices. A live video can be formed. Among them, because the server always generates a new slice of the latest live data, the client only needs to continuously play the slice obtained from the server in order, and the live broadcast is realized.

There are currently two ways to generate slices that conform to the HLS protocol: one way is more The transcoding process of the media data directly generates the slice that meets the requirement; the second method is to separate the transcoding and the fragmentation process separately, first transcoding the live data, and then slicing the transcoded file; Among them, the first method requires that the transcoder must be able to support the slicing function, which has high design requirements for the transcoder, which is not conducive to implementation; the second method is easier to implement because it converts the transcoding and fragmentation independently. . Therefore, the second method of generating slices is more popular, but it also has the following problems: due to the fixed length of the generated slice, there may be incomplete data or redundant data at the edge of the slice, resulting in a live view of the slice. There is a discontinuity in the live picture between the slow and adjacent two slices, which reduces the user's viewing experience.

Summary of the invention

In order to solve the existing technical problems, the embodiment of the present invention provides a method, a server, and a computer storage medium for acquiring a live video slice, which can solve the slow occurrence of the live broadcast picture of the slice, and the live picture between the adjacent two slices exists. Discontinuous conditions, etc., thereby improving the smoothness of live video playback, the quality of the playback picture, and improving the user viewing experience.

The technical solution of the embodiment of the present invention is implemented as follows:

An embodiment of the present invention provides a method for obtaining a live video slice, where the method includes:

Get the baseline duration of the slice;

Determining a sum of a reference duration and a first preset duration as a first duration, and/or determining a difference between the reference duration and the first preset duration is a second duration;

Get the qth key frame of the current original live broadcast source;

Determining a starting position of the qth key frame as a starting position of the qth slice;

Determining that the starting position of the q+1th key frame is the first time when the q+1th key frame exists in the first duration and/or the second duration of the start position of the qth key frame The end position of q slices;

At the start position and the end position of the qth frame, the current original live broadcast source is segmented to form a qth slice; wherein q is a positive integer.

In the above solution, the method further includes:

In the first duration and/or the second duration of the start position of the qth key frame, when it is determined that there is no q+1th key frame, the position from the start position of the qth key frame is the reference duration At the end position of the qth slice, at the start position and the end position of the qth frame, the current original live broadcast source is segmented to form a qth slice.

In the above solution, the reference duration for obtaining the slice includes:

Get the key frame length of the current original live broadcast source;

Obtaining a video buffer checker VBV length for buffering the current original live broadcast source;

Obtaining a sum of a key frame length, a VBV length, and a first preset threshold;

The sum value obtained is taken as the reference duration of the slice.

In the above solution, before acquiring the reference duration of the slice, the method includes:

Receiving the current original live broadcast source;

Converting the current original live broadcast source into a real-time transport protocol RTP stream, and identifying key frame information of the RTP stream;

Cache, sort, and parse the RTP stream;

Deleting the RTP header in the processed RTP stream to form RTP payload data;

In the RTP payload data, the qth key frame is obtained according to the key frame information.

In the above solution, the method further includes:

Obtaining the program association table PAT and the program mapping table PMT from the current original live broadcast source, and configuring the PAT and the PMT to the initial position of each slice to facilitate the playback of the program content carried by the slice;

When there are at least two code rates in the current live broadcast source, the lengths of the program content carried by the slices configured with different code rates are the same, so as to facilitate the switching of different code rate slices.

The embodiment of the invention further provides a server, the server comprising:

a first obtaining unit configured to acquire a reference duration of the slice;

The first determining unit is configured to determine that the sum of the reference duration and the first preset duration is the first duration, And/or determining a difference between the reference duration and the first preset duration is a second duration;

a second acquiring unit, configured to acquire a qth key frame of the current original live broadcast source;

a second determining unit, configured to determine that a starting position of the qth key frame is a starting position of the qth slice;

a third determining unit, configured to determine, when the q+1th key frame exists in the first duration and/or the second duration of the start position of the qth key frame, the q+1th key The starting position of the frame is the end position of the qth slice;

The first sub-unit is configured to split the current original live broadcast source at the start position and the end position of the q-th frame to form a q-th slice; wherein q is a positive integer.

In the above scheme,

The third determining unit is further configured to determine that the qth key is the distance when there is no q+1th key frame in the first duration and/or the second duration of the qth key frame start position. The position where the frame start position is the reference duration is the end position of the qth slice;

Correspondingly, the first segmentation unit is further configured to slice the current original live broadcast source at the start position and the end position of the qth frame to form a qth slice.

In the foregoing solution, the first acquiring unit is further configured to:

Get the key frame length of the current original live broadcast source;

Obtaining a video buffer checker VBV length for buffering the current original live broadcast source;

Obtaining a sum of a key frame length, a VBV length, and a first preset threshold;

The sum value obtained is taken as the reference duration of the slice.

In the above solution, the server further includes:

The first receiving unit is configured to receive the current original live broadcast source;

The first converting unit is configured to convert the current original live broadcast source into a real-time transport protocol RTP stream, and identify key frame information of the RTP stream;

a first processing unit, configured to cache, sort, and parse the RTP stream;

a first deleting unit, configured to delete an RTP header in the RTP stream processed by the first processing unit to form RTP payload data;

corresponding,

The second acquiring unit is further configured to acquire, in the RTP payload data, the qth key frame according to the key frame information.

In the above solution, the server further includes:

a third obtaining unit, a first configuration unit, a fourth determining unit, and a second configuration unit; wherein

The third obtaining unit is configured to obtain a program association table PAT and a program mapping table PMT from the current original live broadcast source;

The first configuration unit is configured to configure PAT and PMT to an initial position of each slice to facilitate playing of the program content carried by the slice;

The fourth determining unit, configured to determine that the current original live broadcast source has at least two code rates, triggering the second configuration unit;

The second configuration unit is configured to configure the lengths of the program content carried by the slices of different code rates to be the same, so as to facilitate switching of different code rate slices.

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the foregoing method for obtaining a live video slice.

The method for obtaining a live video slice, the server, and the computer storage medium provided by the embodiment of the present invention include: obtaining a reference duration of the slice; determining a sum of the reference duration and the first preset duration as a first duration, and/or determining The difference between the reference duration and the first preset duration is the second duration; obtaining the qth key frame of the current original live broadcast source; determining the starting position of the qth keyframe as the starting position of the qth slice; Determining that the q+1th key frame is the qth slice when the q+1th key frame exists in the first duration and/or the second duration of the start position of the key frame End position; at the beginning and end of the qth frame, the current original straight The broadcast source performs segmentation to form a qth slice; wherein q is a positive integer. With the technical solution of the embodiment of the present invention, the slice length can be fluctuated around the reference duration, and the live broadcast picture of the slice can be slowed down, and the live picture between the adjacent two slices is discontinuous, thereby improving the live broadcast. The smoothness of video playback and the quality of the playback picture enhance the user's viewing experience.

DRAWINGS

FIG. 1 is a schematic flowchart of a method for obtaining a live video slice according to an embodiment of the present invention;

2 is a schematic diagram of an application scenario according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a specific implementation process of a method for obtaining a live video slice according to an embodiment of the present disclosure;

4 is a schematic diagram of an application scenario 2 according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of a server according to an embodiment of the present invention.

detailed description

The preferred embodiments of the present invention are described in detail below with reference to the accompanying drawings.

FIG. 1 is a schematic flowchart of a method for obtaining a live video slice according to an embodiment of the present invention; as shown in FIG. 1 , the method includes:

Step 11: Obtain a reference duration of the slice;

Here, the network side, specifically the network side server receives the original live source stream, such as the MPEG-TS stream sent by the live broadcast source MPEG system, because the original live source stream is continuously sent to the server by the live source system, so the server pair The processing of the original live source stream received each time is the same, please see the description below.

The server converts the MPEG-TS code stream into a Real-time Transport Protocol (RTP) code stream, and identifies key frame information in the RTP code stream, and performs processing such as buffering, sorting, and parsing the RTP stream; After the RTP header of the RTP stream processed as described above, RTP payload data; wherein the key frame information includes: a key frame header, a frame tail; a key frame length in the RTP stream, and a video buffer verifier for buffering the code stream (VBV, Video Buffering Verifier) The length of the key frame length, the VBV length, and the first threshold are summed, and the summation result is used as the reference duration of the slice, that is, if the key frame length is KP, the VBV length is V, and the first threshold is A, then the base time of the slice is BS=KP+V+A.

In the process of buffering, sorting, parsing, etc. of the RTP stream, if the original live source stream has been scrambled (descrambled), the scheme also needs to be scrambled (scrambled); if it is divided into slices later It is also necessary to encrypt the slice. In this solution, encryption processing is also required, which is processed according to the specific application.

The first threshold A is preset, and may be set according to the actual length of the MPEG-TS code stream. For example, A may be set to 1 in advance, or may be set to 2 in advance; the reference duration and key are The unit of frame length, V, and A can be seconds s, minutes min, hour hours, etc., and is usually taken as seconds.

The key frame may also be referred to as an I frame; the purpose of identifying key frame information in the RTP code stream is illustrated in step 13.

Step 12: determining that the sum of the reference duration and the first preset duration is the first duration, and/or determining that the difference between the reference duration and the first preset duration is the second duration;

Here, the first preset duration is set according to the actual application situation; taking the first preset duration as the duration B as an example, calculating the sum of the reference durations BS and B, and taking the sum of the two as the first duration; / or, calculate the difference between the base duration BS and B, and use the difference between the two as the second duration.

Step 13: Obtain the qth key frame of the current original live broadcast source.

Here, based on the key frame information made to the RTP stream, the qth key frame is identified; wherein q is a positive integer.

Step 14: determining that the starting position of the qth key frame is the starting position of the qth slice;

Step 15: Determine, in the first duration and/or the second duration of the start position of the qth key frame, that the q+1th key frame exists, determine the start of the q+1th key frame The position is the end position of the qth slice;

Here, in the first duration and/or the second duration of the start position of the qth key frame, it is searched whether there is a q+1th key frame, and when found, the q+1th key frame is determined. The starting position is the end position of the qth slice.

Step 16: At the start position and the end position of the qth frame, the current original live broadcast source is segmented to form a qth slice; wherein q is a positive integer.

Here, the q-th slice is formed at the start position and the end position of the qth key frame to form a q-th slice; that is, the start position of the q-th slice is the q-th key frame start position, the qth The end position of the slice is the starting position of the q+1th key frame.

In one embodiment, within the first duration and/or the second duration of the start position of the qth key frame, it is determined that the qth key frame start position is not present when the q+1th key frame is not present. The position of the reference duration is the end position of the qth slice, and at the start position and the end position of the qth frame, the current original live broadcast source is segmented to form a qth slice.

For example, the base duration BS=10s and the first preset duration B=1s are taken as an example, the first duration is 9s and the second duration is 11s; in the RTP payload data, according to the key frame made to the RTP stream The information identifies the first key frame, and the start position of the first key frame is the start position of the first slice of the current original live broadcast source, and the start position of the first key frame is 0s, from the 0s position. Find the time to 9s-11s. In the duration of the 9s-11s, when the second key frame is determined, the start position of the second key frame is taken as the end position of the first slice. The start position and the end position of the slice are segmented to form the first slice of the current original live broadcast source; meanwhile, the start position of the second key frame is used as the start position of the second slice of the current original live broadcast source. The starting position of the second slice is 0s, and then searching from the 0s position to the time of 9s-11s. During the duration of the 9s-11s, it is determined that there is a third key frame, and the third key frame is present. The starting position is used as the ending position of the second slice, and is segmented at the start position and the end position of the second slice to form a second slice of the current original live broadcast source; and so on, the server pairs each receive The original live source to the slice is sliced. That is to say, in the RTP payload data, the starting position (the beginning position) of the qth key frame is the starting position (the beginning position) of the qth slice, and the starting position of the q+1th key frame is the The end position of q slices.

It can be seen that in the above solution, a key frame can be divided into the same slice. In this solution, the position of the key frame is the segmentation position, or the distance from the q-th key frame start position is used as the reference duration. The location is the segmentation location, and the original live source stream is segmented to ensure that the HLS slice length fluctuates around the reference duration, so that the key frames are sliced into the same slice, which solves the incomplete data or data redundancy of the slice edge. The phenomenon of the live broadcast of the slice is slow, and the live picture between the adjacent two slices is discontinuous, which improves the smoothness of the live video playback, the quality of the playback picture, and improves the user viewing experience.

When slicing the original live source stream, the sliced slice also needs to be satisfied:

1) Obtain a program association table (PAT, Program Association Table) and a program map table (PMT) from the live source stream, and configure PAT and PMT at the initial position of each slice to facilitate slicing the program content carried. Play

The PAT table mainly includes a program channel number and a packet identifier (PID, Packet Identifier) number of the PMT corresponding to each channel; the PMT table includes PIDs of all video streams of each channel, and audio data of all audio streams. The PID and other PIDs; when the terminal requests to play a certain video of the current channel, the Video data can be filtered from the channel through the PID of the Video; meanwhile, the PAT and the PMT periodically appear in the original live source stream. In order to facilitate the terminal to access the code stream of the played video at any time when requesting to play the video;

For the description of the PAT and the PMT, refer to the related descriptions, and no further details are provided here. In this scheme, PAT and PMT are configured at the beginning of each slice, enabling independent playback of each slice.

2) When it is determined that there are at least two code rates in the current original live broadcast source, the lengths of the program content carried by the slices configured with different code rates are the same, so as to facilitate switching of different code stream slices, that is, in order to facilitate live video Switch between different code rates.

Since the HLS protocol, each slice (also referred to as a TS slice) has its own supported play rate, sequence number, formation time, etc., so by looking up the index file (m3u8 format), it can be found All slices of the same play rate, the slices are sorted according to the size of the slice number.

For example, taking the original live source stream as a single code rate (1024 kbps), the live single rate file includes: a main index file, a sub-index file, and a TS slice; wherein the main index file is named index.m3u8, and the sub-index file is The code rate is named 1024.m3u8 (the code rate is 1024 kbps), and the TS slice is named $time-1024-$seq.ts; where: $time indicates the slice formation time, 1024 indicates the slice rate is 1024 kbps, and $seq indicates the slice. The serial number; the slices with the same playback rate need to be placed under the same sub-index file. At the same time, the file content of the sub-index file 1024.m3u8 can represent the names of all the slices that it can index. For example, when the sub-index file 1024.m3u8 can be indexed by up to 3 TS slices, it can be indexed to Slice 1 of 20140521160000-1024-1.ts, slice 2 named 20140521160000-1024-2.ts, and slice 3 named 20140521160000-1024-2.ts. The original live source stream is multi-code rate similar to the description of the original live source stream as the single code rate, except that in the multi-rate original live source stream, the number of sub-index files is equal to the code used in the original live source stream. The number of rates.

Since the embodiment of the present invention relates to a live multimedia code stream, as the server continuously acquires the original live source stream, the original live stream is continuously segmented, and new slices are continuously formed, so the main and sub index files are required to be The new slice is continuously updated. For example, within 1 min, the server can split the original live source stream into 6 TS slices, then the sub-index file is updated every 10s, updated to index the 6 latest TS slices, and deleted. For the previous 6 TS slices The index is adapted to the requirements of the HLS live specification; this process can be seen as a rollback process for slices.

In the above solution, for the specific rollback of the slice and the specific process of finding a slice for the primary and secondary index files, refer to the related description of the HLS protocol, and details are not described herein.

2 is a schematic diagram of an application scenario according to an embodiment of the present invention; FIG. 3 is a schematic flowchart of a specific implementation of a method for obtaining a live video slice according to an embodiment of the present invention; and FIG. 2 and FIG. 3 further describe an embodiment of the present invention.

In FIG. 2, the server is located in a single node, the original live source stream is a single code rate, the first threshold is A=1s, the first preset duration B=1s, the first duration is 9s, and the second duration is 11s. :

Step 301: The node, specifically the server located at the node, receives the original live broadcast source MPEG-TS code stream sent by the MPEG system;

The server supports a User Data Protocol (UDP) unicast and UDP multicast of a code stream.

Step 302: The server converts the received MPEG-TS code stream into an RTP code stream.

Here, the server will add an extended RTP header to the MPEG-TS code stream, determine the length of the VBV, and the key frame information of the code stream such as the key frame header, the end of the frame, the length, etc.; obtain the PAT and PMT of the code stream.

Step 303: The server performs processing such as buffering, sorting, parsing, and the like on the RTP stream to facilitate playback of the subsequent terminal.

Here, after data caching, sorting, parsing, etc., the data may be scrambled, descrambled, encrypted, etc., depending on the specific application.

Step 304: Determine a reference duration BS;

Here, BS = key frame length KP + VBV length V + 1.

Step 305: Remove the header of the RTP stream to obtain RTP payload data.

Step 306: In the RTP payload data, obtain the qth key frame of the current original live broadcast source; determine the starting position of the qth key frame as the starting position of the qth slice; Finding whether there is a q+1th key frame in the first duration and/or the second duration of the start position of the key frame;

When it is found, step 307 is performed; otherwise, step 308 is performed.

Step 307: Determine the start position of the q+1th key frame as the end position of the qth slice; at the start position and the end position of the qth frame, segment the current original live broadcast source to form the qth Slice, continue to step 309;

Here, in the RTP payload data, the beginning position of the qth key frame is the beginning position of the qth slice, and the beginning position of the q+1th key frame is the end position of the qth slice; I will not repeat them here.

Step 308: The position at which the q-th key frame start position is the reference duration is the end position of the q-th slice, and at the start position and the end position of the q-th frame, the current original live broadcast source is segmented to form a first q slices, continue to step 309;

In one embodiment, after the RTP payload data is sliced into slices, the PAT and PMT of the code stream need to be placed at the beginning of each slice so that independent playback of each slice can be achieved.

Step 309: The server generates a sub-index file for the formed slice according to the HLS protocol.

Here, since the original live broadcast source is a single code rate, there is only one sub index file.

Step 310: The server refreshes the formed slice, the sliced primary index file, and the sub-index file to the disk for storage.

Here, upon receiving the MPEG-TS code stream, the server generates a main index file in accordance with the HLS protocol.

Step 311: When the terminal reports a video play request for the current 10s of the channel A, the terminal requests the main index file from the download server (DLS, Down Load Server) in the node; the DLS reads the main index file of the channel A in the disk. And sending the file content to the terminal through the HTTP protocol; the terminal requests the sub-index file from the DLS; the DLS reads the sub-index file corresponding to the video within the current 10s of the channel A in the disk, and sends the file content to the terminal through the HTTP protocol. ; terminal to DLS please Finding a TS slice file corresponding to the sub-index file; the DLS reads the TS slice file of the current 10s video carrying the channel A in the disk, and sends the file content to the terminal through the HTTP protocol, and the terminal plays the TS slice for the User watch.

FIG. 4 is a schematic diagram of an application scenario 2 according to an embodiment of the present invention; as shown in FIG. 4, the server in the solution may also be applied to a multi-node scenario; wherein each node may be connected to the terminal to facilitate the terminal request. Live data, each node includes a server and a download server DLS; further, the central node, specifically the server of the central node receives the MPEG-TS code stream, and transmits the MPEG-TS code stream to the regional node, specifically the regional node The server of the area node transmits the MPEG-TS code stream to the edge node; after the corresponding server in the three nodes receives the live source source stream, the processing is performed according to the foregoing steps 302 to 311, and details are not described herein again.

The embodiment of the invention further provides a computer storage medium, wherein the computer storage medium stores computer executable instructions, and the computer executable instructions are used in the foregoing method for obtaining a live video slice.

Based on the foregoing method for obtaining a live video slice, the embodiment of the present invention further provides a server, and FIG. 5 is a schematic diagram of a server according to an embodiment of the present invention; as shown in FIG. 5, the server includes: a first acquiring unit 500. a first determining unit 501, a second obtaining unit 502, a second determining unit 503, a third determining unit 504, and a first splitting unit 505;

The server further includes a first receiving unit, a first converting unit, a first processing unit, and a first deleting unit; wherein

The first obtaining unit 500 is configured to acquire a reference duration of the slice;

Here, the server, specifically the first receiving unit, receives the original live source stream, such as receiving an MPEG-TS stream from the MPEG system, and transmits the received MPEG-TS stream to the first converting unit; The converting unit converts the MPEG-TS code stream into an RTP code stream, and identifies key frame information in the RTP code stream; the first processing unit buffers, sorts, parses, etc. the RTP code stream; the first deleting unit deletes the first frame Processing unit processing RTP code stream The RTP packet header forms the RTP payload data; the first obtaining unit 500 obtains the key frame length and the VBV length in the RTP code stream, and sums the key frame length, the VBV length, and the first threshold, and the summation result is used as a slice. The reference duration, that is, if the key frame length is KP, the VBV length is V, and the first threshold is A, the first obtaining unit 500 determines the reference duration of the slice BS=KP+V+A. The first processing unit in the process of buffering, sorting, parsing, etc. processing the RTP stream, if the original live source stream has been scrambled (descrambled), the first processing unit in the solution needs to perform descrambling ( Scrambling) operation; if the slice needs to be encrypted when it is subsequently divided into slices, the first processing unit also needs to perform encryption processing.

The first determining unit 501 is configured to determine that the sum of the reference duration and the first preset duration is a first duration, and/or determine that the difference between the reference duration and the first preset duration is a second duration;

Here, taking the first preset duration as the duration B, the first determining unit 501 calculates the sum of the reference durations BS and B, and uses the sum of the two as the first duration; and/or calculates the reference durations BS and B. The difference between the two is taken as the second duration.

The second obtaining unit 502 is configured to acquire the qth key frame of the current original live broadcast source;

Here, the second obtaining unit 502 identifies the qth key frame according to the key frame information in the RTP payload data.

a second determining unit 503, configured to determine that a starting position of the qth key frame is a starting position of the qth slice;

The third determining unit 504 is configured to determine, when the q+1th key frame exists in the first duration and/or the second duration of the qth key frame start position, the q+1th The starting position of the key frame is the end position of the qth slice;

Here, the third determining unit 504 searches for the q+1th key frame in the first duration and/or the second duration of the qth key frame start position, and when found, determines the q+th The starting position of one key frame is the end position of the qth slice.

The first unit 505 is configured to be at the start position and the end position of the qth frame, for the current The original live broadcast source is segmented to form a qth slice; wherein q is a positive integer.

Here, the first slicing unit 505 performs segmentation at the qth key frame start position and end position to form a qth slice; that is, the qth slice starts from the qth key frame. The starting position, the end position of the qth slice is the q+1th key frame starting position.

The third determining unit 504 is further configured to: when the q+1th key frame is not present, the qth distance is determined in the first duration and/or the second duration of the qth key frame start position. The position where the key frame start position is the reference duration is the end position of the qth slice;

Correspondingly, the first segmentation unit 505 is further configured to segment the current original live broadcast source at the start position and the end position of the qth frame to form a qth slice.

For example, the base duration BS=10s and the first preset duration B=1s are taken as an example, the first duration is 9s and the second duration is 11s; in the RTP payload data, the second obtaining unit 502 is based on the RTP code. The key frame information of the stream identifies the first key frame, and the second determining unit 503 determines that the start position of the first key frame is the start position of the first slice of the current original live broadcast source, and the third determining unit 504 The starting position of the first key frame is 0s, and the position from the 0s position is searched to the time of 9s-11s. During the period of 9s-11s, it is determined that there is a second key frame, and the second key frame is The start position is the end position of the first slice, and the first segmentation unit 505 performs segmentation at the start position and the end position of the first slice to form the first slice of the current original live broadcast source; meanwhile, the second The determining unit 503 determines that the starting position of the second key frame is the starting position of the second slice of the current original live broadcast source, and the third determining unit 504 searches for the second slice starting position as 0s, and then searches from the 0s position. Up to the 9s-11s duration, within the duration of the 9s-11s, it is determined that there is a third key In the frame, the start position of the third key frame is taken as the end position of the second slice, and the first segmentation unit 505 performs segmentation at the start position and the end position of the second slice to form the current original live broadcast source. The second slice; and so on, the slice of the original live feed received each time.

It can be seen that in the above solution, the beginning position of the qth key frame is determined to be the qth slice. The starting position (the beginning position) determines that the starting position (the beginning position) of the q+1th key frame is the ending position of the qth slice.

The server further includes:

a third acquisition unit, a first configuration unit, a fourth determination unit, and a second configuration unit (not illustrated in FIG. 5);

The third obtaining unit is configured to obtain a program association table PAT and a program mapping table PMT from the current original live source stream;

The first configuration unit is configured to configure PAT and PMT to an initial position of each slice to facilitate playing of the program content carried by the slice;

The PAT table mainly includes a program channel number and a PID of the PMT corresponding to each channel, and a code; the PMT table includes PIDs of all video streams of each channel, PIDs of all audio streams, and other PIDs; When the terminal requests to play a certain video of the current channel, the Video data can be filtered out from the channel through the PID of the Video; meanwhile, the PAT and the PMT periodically appear in the original live source stream to facilitate the terminal to request playback. The code stream of the played video can be accessed at any time during the video;

The fourth determining unit, configured to determine that the current original live broadcast source has at least two code rates, triggering the second configuration unit;

The second configuration unit is configured to configure the lengths of the program content carried by the slices of different code rates to be the same, so as to facilitate switching of different code rate slices.

In this scheme, PAT and PMT are configured at the beginning of each slice, enabling independent playback of each TS slice.

In the above scheme, the original broadcast source stream is segmented by taking the position of the key frame as the segmentation position or the position of the q-th key frame starting position as the reference position, and ensuring the HLS slice length is The base time fluctuates around the reference time, so that the key frame is split into the same slice, which solves the problem that the data of the slice edge is incomplete or the data is redundant, thereby solving the sliced The live broadcast screen appears slower, and the live video between adjacent two slices has discontinuous problems, which improves the smoothness of live video playback, the quality of the playback picture, and improves the user viewing experience.

Those skilled in the art should understand that the implementation functions of the processing units in the server shown in FIG. 5 can be understood by referring to the related description of the foregoing method for obtaining live video slices. Those skilled in the art should understand that the functions of the processing units in the server shown in FIG. 5 can be implemented by a program running on the processor, or can be implemented by a specific logic circuit.

In a practical application, the first obtaining unit 500, the first determining unit 501, the second obtaining unit 502, the second determining unit 503, the third determining unit 504, and the first splitting unit 505 may each be configured by a central processing unit (CPU). Central Processing Unit), or a digital signal processing (DSP), or a microprocessor (MPU, Micro Processor Unit), or a Field Programmable Gate Array (FPGA); The CPU, DSP, MPU, and FPGA can all be built into the node, specifically the server.

Those skilled in the art will appreciate that embodiments of the present invention can be provided as a method, system, or computer program product. Accordingly, the present invention can take the form of a hardware embodiment, a software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) including computer usable program code.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

The above is only the preferred embodiment of the present invention and is not intended to limit the scope of the present invention.

Industrial applicability

An embodiment of the present invention provides a method, a server, and a computer storage medium for acquiring a live video slice, where the location of the key frame is a segmentation position, or the location of the q-th key frame starting position is a segmentation time. The location, the original live source stream is segmented, which can ensure that the HLS slice length fluctuates around the reference duration, so that the key frame is split into the same slice, which solves the problem that the slice edge has incomplete data or data redundancy, thereby solving the problem. The live broadcast screen of the slice appears slower, and there are problems such as discontinuity in the live video between the adjacent two slices, which improves the smoothness of the live video playback, the quality of the playback picture, and improves the user viewing experience.

Claims (11)

A method for obtaining a live video slice, the method comprising: Get the baseline duration of the slice; Determining a sum of a reference duration and a first preset duration as a first duration, and/or determining a difference between the reference duration and the first preset duration is a second duration; Get the qth key frame of the current original live broadcast source; Determining a starting position of the qth key frame as a starting position of the qth slice; Determining that the starting position of the q+1th key frame is the first time when the q+1th key frame exists in the first duration and/or the second duration of the start position of the qth key frame The end position of q slices; At the start position and the end position of the qth frame, the current original live broadcast source is segmented to form a qth slice; wherein q is a positive integer. The method of claim 1 wherein the method further comprises: In the first duration and/or the second duration of the start position of the qth key frame, when it is determined that there is no q+1th key frame, the position from the start position of the qth key frame is the reference duration At the end position of the qth slice, at the start position and the end position of the qth frame, the current original live broadcast source is segmented to form a qth slice. The method of claim 1, wherein the obtaining a reference duration of the slice comprises: Get the key frame length of the current original live broadcast source; Obtaining a video buffer checker VBV length for buffering the current original live broadcast source; Obtaining a sum of a key frame length, a VBV length, and a first preset threshold; The sum value obtained is taken as the reference duration of the slice. The method according to any one of claims 1 to 3, wherein the reference of the slice is obtained Before the duration, the method includes: Receiving the current original live broadcast source; Converting the current original live broadcast source into a real-time transport protocol RTP stream, and identifying key frame information of the RTP stream; Cache, sort, and parse the RTP stream; Deleting the RTP header in the processed RTP stream to form RTP payload data; In the RTP payload data, the qth key frame is obtained according to the key frame information. The method of claim 4 wherein the method further comprises: Obtaining the program association table PAT and the program mapping table PMT from the current original live broadcast source, and configuring the PAT and the PMT to the initial position of each slice to facilitate the playback of the program content carried by the slice; When there are at least two code rates in the current live broadcast source, the lengths of the program content carried by the slices configured with different code rates are the same, so as to facilitate the switching of different code rate slices. A server, wherein the server comprises: a first obtaining unit configured to acquire a reference duration of the slice; The first determining unit is configured to determine that the sum of the reference duration and the first preset duration is a first duration, and/or determine that the difference between the reference duration and the first preset duration is a second duration; a second acquiring unit, configured to acquire a qth key frame of the current original live broadcast source; a second determining unit, configured to determine that a starting position of the qth key frame is a starting position of the qth slice; a third determining unit, configured to determine, when the q+1th key frame exists in the first duration and/or the second duration of the start position of the qth key frame, the q+1th key The starting position of the frame is the end position of the qth slice; The first sub-unit is configured to split the current original live broadcast source at the start position and the end position of the q-th frame to form a q-th slice; wherein q is a positive integer. The server according to claim 6, wherein The third determining unit is further configured to determine that the qth key is the distance when there is no q+1th key frame in the first duration and/or the second duration of the qth key frame start position. The position where the frame start position is the reference duration is the end position of the qth slice; Correspondingly, the first segmentation unit is further configured to slice the current original live broadcast source at the start position and the end position of the qth frame to form a qth slice. The server according to claim 6, wherein the first obtaining unit is further configured to: Get the key frame length of the current original live broadcast source; Obtaining a video buffer checker VBV length for buffering the current original live broadcast source; Obtaining a sum of a key frame length, a VBV length, and a first preset threshold; The sum value obtained is taken as the reference duration of the slice. The server according to any one of claims 6 to 8, wherein the server further comprises: The first receiving unit is configured to receive the current original live broadcast source; The first converting unit is configured to convert the current original live broadcast source into a real-time transport protocol RTP stream, and identify key frame information of the RTP stream; a first processing unit, configured to cache, sort, and parse the RTP stream; a first deleting unit, configured to delete an RTP header in the RTP stream processed by the first processing unit to form RTP payload data; corresponding, The second acquiring unit is further configured to acquire, in the RTP payload data, the qth key frame according to the key frame information. The server of claim 9, wherein the server further comprises: a third obtaining unit, a first configuration unit, a fourth determining unit, and a second configuration unit; wherein The third obtaining unit is configured to obtain a program association table PAT from the current original live broadcast source. And program map table PMT; The first configuration unit is configured to configure PAT and PMT to an initial position of each slice to facilitate playing of the program content carried by the slice; The fourth determining unit, configured to determine that the current original live broadcast source has at least two code rates, triggering the second configuration unit; The second configuration unit is configured to configure the lengths of the program content carried by the slices of different code rates to be the same, so as to facilitate switching of different code rate slices. A computer storage medium having stored therein computer executable instructions for performing the method of any one of claims 1 to 5.

Images