WO2020061622A1

WO2020061622A1 - A storybook compilation system

Info

Publication number: WO2020061622A1
Application number: PCT/AU2019/051016
Authority: WO
Inventors: Derek Barker; Alan Clarke
Original assignee: Oobee Doobee And Friends Pty Ltd
Current assignee: Oobee Doobee And Friends Pty Ltd
Priority date: 2018-09-26
Filing date: 2019-09-23
Publication date: 2020-04-02
Anticipated expiration: 2021-03-26

Abstract

A storybook compilation system has a database having storybook templates, each having a plurality of page templates. At least one page template has page image data and defines at least one placeholder region. A storybook selection interface allows for the the selection of a selected storybook template and an image upload interface allows for uploading image data comprising facial data. A merge controller merges facial image data with the page image data at the at least one placeholder region for generating compiled storybook data.

Description

A storybook compilation system

Field of the I nvention

[1] This invention relates generally to a storybook com pilation system configured to generate personalised storybook data in accordance with uploaded image data .

Summa ry of the Disclosure

[2] There is provided herein a storybook com pilation system for the automated generation of personal ised storybook data in accordance with upl oaded image data.

[3] Specifical ly, the system com prises a database com prising a plural ity of storybook tem plates, each having a pl ural ity of page tem plates having at least one placeholder region.

[4] A user may select a selected storybook using a storybook select ion interface provided by the system and upload image data of an image incl uding a face. The system com prises a merge control ler which is configured for merging facial image data with the page image data at the at least one placeholder region for generatin g com piled storybook data.

[5] The system may com prise an image recognition control ler which identifies facial regions within the uploaded image from which the facial image data may be extracted by the merge controller for merging. Alternatively, the image recognition control ler may identify facial attributes which are used to custom ise a character avatar for merging.

[6] The system may com prise an image manipulation control ler to adjust extracted facial image data such as by adjusting tonal and/or intensity levels, including in accordance with surrounding image data of the page tem plate. The image manipulation controller may further perform gaze and/or tilt orientation adjustment to most closely match those of the placeholder regions.

[7] The at least one placeholder region may define a gaze orientation a nd the image recognition control ler may configured for identifying a pl urality of facial regions within the uploaded image data . As such, the merge controller may select the most optimal facial region according to the defined gaze orientation.

[8] The merge controller may extract the facial image data from the uploaded image data at the regions identified by the image recognition control ler. The merge control ler may perform facial boundary detection to extract pixels falling within facial boundaries. The merge control ler may perform hair region boundary detection to exclude pixels falling within hair boundaries. [9] The placeholders may define different characters and the image recognition controller may further perform subject identity recognition (such as child and grandparent subjects) for extraction of separate facial image data for the separate subjects.

[10] The system may serve a character type selection interface for the selection of a character type. As such, the merge control ler may furthermore merge character type image data at the at least one placeholder region according to the selected character type . The page tem plate data may com prise a pl ural ity of layers such that the merge control ler may merge the character type image data seam lessly between foreground and background layers.

[11] In em bodiments, the at least one placeholder region may specify a character type poise, such as run ning, jum ping, sitting poises and the like. As such, the merge control ler may furthermore select the character type image data according to the specified poise.

[12] In accordance with one aspect, there is provided a storybook com pilation system com prising: a database com prising a plural ity of storybook tem plates, each com prising a plural ity of page tem plates, wherein at least one page tem plate has page image data and defines at least one placeholder region; a storybook selection interface for the selection of a selected storybook tem plate; an image upload interface for uploading image data com prising an image including a face; and a merge controller which merges facial image data derived from the image data with the page image data at the at least one placeholder region for generating com piled storybook data.

[13] The system may serve an information data field interface com prising a pl urali ty of data fields for receipt of associated data field values and the merge control ler may be configured for merging the data field values with the page image data.

[14] At least one page tem plate may define at least one data field placeholder region and the system may be configured for dynam ically configuring the information data field interface with input data fields according to meta data associated with the at least one data field placeholder region.

[15] The at least one page tem plate may define a plurality of layers and the merge controller may be configured for merging the facial data at one of the layers.

[16] The system may be configured for serving a character type selection interface for the receipt of a selected character ty pe and the merge control ler may be configured for merging character type image data with the page image data according to the selected character type with the page image data. [17] The system may further com prise an image recognition controller and the image recognition controller may be configured for identifying at least one attribute from the uploaded image data and for selecting a character type accordingly.

[18] The at least one placeholder region may have associated meta data defining a character poise and the merge controller may be configured for selecting the character type image data according to the defined character poise.

[19] The system may be configured for serving a character augment selection interface for the receipt of a selected character augment and th e merge control ler may be configured for merging the character type image data with augment image data according to the selected character augment.

[20] The system may further com prise an image recognition controller which performs facial image recognition on the uploaded image data to identify a facial region within the uploaded image data.

[21] The merge control ler may be configured for extracting the facial image data from the uploaded image data at the identified facial region.

[22] The image recognition controller may be configured for identifying a facial boundary and the merge control ler may be configured for extracting pixels from the uploaded image data within the facial boundary.

[23] The image recognition controller may be configured for identifying a hairl ine boundary and the merge control ler may be configured for excl uding pixels from the uploaded image data within the hairl ine boundary.

[24] The merge controller may be configured for perform ing at least one of tonal and intensity adjustments on pixels extracted from the uploaded image data.

[25] The at least one of the total and intensity adjustments are performed in accordance with at least one of a respective tonal and intensity level of image data of the page tem plate.

[26] The system may further com prise an image recognition controller configured to identify at least facial attribute from the uploaded image data com prising at least one of hair colour and eye colour and the merge control ler may be configured for augmenting a character avatar according to the at least one facial attribute and facial image data may be derived from the augmented character avatar.

[27] The image may com prise a plurality of faces and the merge control ler may be configured for selecting an optimal facial region therefrom . [28] The at least one placeholder region may have metadata defining a gaze orientation and the merge controller may be configured for selecting the optimal facial region in accordance with the defined gaze orientation.

[29] The image upload interface may be configured for indicating that the uploaded image data may com prise insufficient faces to satisfy defined gaze orientations of the at least one placeholder region meta data.

[30] The at least one placeholder region may define a gaze orientation a nd the merge controller may be configured to map two -dimensional facial image data extracted from the uploaded image data onto a 3D facial model, rotate the 3D model according to the gaze orientation and extract 2 D gaze adjusted facial image data therefrom for merging.

[31] The page tem plates may define at least two p laceholder regions having associated meta data defining at least two different character types and the image recognition controller may be further configured for identity recognition to identify faces of at least two separate subjects from the uploaded image data and the merge control ler merges separate facial image data extracted from the uploaded image data for each of the at least two subjects at each of the at least two respective placeholder regions.

[32] Other aspects of the invention are also disclosed.

Brief Description of the Drawings

[33] Notwithstanding any other forms which may fal l within the scope of the present invention, preferred em bodiments of the disclosure wil l now be described, by way of exam ple only, with reference to the accom panying drawings in which :

[34] Figure 1 shows a storybook com pilation system in accordance with an em bodiment;

[35] Figure 2 shows exem plary processing of the system in accordance with an em bodiment;

[36] Figure 3 shows an exemplary page tem plate in accordance with an em bodiment;

[37] Figure 4 shows exem plary character types and facial placeholder regions in accordance with an em bodiment;

[38] Figure 5 shows an exem plary augmentation interface in accordance with an em bodiment; and

[39] Figure 6 shows an exemplary page generated by the system in accordanc e with an em bodiment. Description of Embodiments

[40] A storybook compiling system 100 may comprise a server 101 in operable communication with at least one client computer terminal 102 across a wide area network 103. The server 101 may comprise a processor 104 processing digital data. In operable communication with the processor 104 via a system bus 105 is a memory device 106. The memory device 106 is configured for storing digital data including computer program code instructions. In use, the processor 104 fetches these computer program code instructions and associated data from the memory device 106 for interpretation and execution of the computer processing functionality provided herein.

[41] The computer program code instructions may be logically divided into a plurality of controllers 107 which act on associated data 108.

[42] The controllers 107 may comprise an image recognition controller 109 for performing image recognition, an image manipulation controller 110 for performing image manipulation, an image augmentation controller 111 for performing image augmentation, a merge controller 112 for merging image and page template data and a compilation controller 113 for generating compiled storybook data.

[43] The data 108 may comprise a plurality of storybook templates, each comprising a plurality of page templates 114. The page templates 114 may comprise at least one facial placeholder region 149.

[44] The data 108 may further comprise a plurality of character type and/or character augments 115. The data 108 may further comprise facial images 116 obtained from uploaded image data. The data 108 may comprise a plurality of compiled storybook pages 117 generated in accordance with the page template data 114 and the facial image data 116 and, in embodiments, the character/augment image data 115.

[45] The memory device 106 may further comprise a plurality of applications 119 including a web server application 120 to serve HTTP web resource requests from the client terminal 102. The server 101 may comprise a network interface 118 for sending and receiving data across a wide area network 103.

[46] The client terminal 102 may similarly comprise a processor 104 in operable communication with a memory device 106 across a system bus 105. The memory device 106 of the client terminal 102 may comprise a web browser application 121 for sending requests to the web server application 120 of the server 101. The client terminal 102 may further comprise a network interface 118 for communicating with the server 101 across the wide area network 103. [47] The client term inal 102 may further com prise an I/O interface 122 including for interfacing with a digita l display device 123 for the display of digital information thereon . The digital display device 123 may be configured for displaying a pl urality of user interfaces thereon . Figure 2 ill ustrates exem plary processing 124 of the system 100.

[48] The processing 124 comprises serving a storybook tem plate selection interface com prising a plural ity of storybook tem plates and receiving a selection of a storybook tem plate at step 125.

[49] The processing 124 may com prise serving an information data field interface and receiving associated data field val ues at step 126. The data fields may com prise fields for a subject's name (incl uding first and last name), a greeting message and the l ike.

[50] Figure 3 shows an exemplary page tem plate 114 in accordance with an em bodiment which is used to generate a generated page 152 as illustrated in Figure 6. The page tem plate 114 may com prise image data 141. The page template 114 may com prise a plurality of layers each having associated image data 141.

[51] The page tem plate 114 may com prise an information placeholder region 142.

[52] In embodiments, the information data field interface is dynam ically configured in accordance with the information placeholder regions 142 of the page tem plates of the selected storybook template. For exam ple, page tem plates 114 of a selected storybook tem plate may com prise a pl urality of information placeholder regions 142 and associated meta data such as, for exam ple, "welcome message", "happy ending message", "a favourite toy of <subject>" and the l ike. As such, the server 101 may dynam ically configure the data fields of the information data field interface to present input fields for the input of a welcome message, a happy ending message and a favour ite toy of the subject as specified by the storybook tem plate.

[53] The processing 124 may com prise the server 101 serving a character type selection interface and receiving a character type selection at step 127. Figure 4 shows an exem plary camouflaged explorer character type 143 com prising, for exam ple, a male character sub- type 144 and a female character sub-type 145. Using the character type selection interface, the user can select a character type, such as, for example, a male explorer character type .

[54] In em bodiments, the server 101 may util ise the image recognition controller 109 to recognise at least one attribute from uploaded image data to automate the selection of the character type accordingly. For exam ple, the image recognition controller 109 may identify a gender attribute so as to automate the selection of the male character type 144 if recognising a male face within the uploaded image data for exam ple. In alternative em bodiments, the server 101 may automate the selection of the character type of by a provided data field value, such as gender.

[55] Other characteristics may be recognised such as an age category including baby, toddler, child, parent, grandparent age categories and the like such that, for exam ple, the server 101 may automate the selection of a toddler character type when the image recognition controller 109 recognises a toddler within the uploaded image data. Sim ilarly, the age category and associated character selection may be made by way of a provided data field value, such as age.

[56] The processing 124 may com prise serving a character augment selection interface 146 and receiving a selected character augment at step 128. Figure 5 il lustrates an exem plary character augment selection interface 146 wherein the character augments represent a pl ural ity of different types of hats 147. Each of the hat augments 147 are selectable and the interface 146 may com prise a live preview 148 which dynam ically displays a facial image augmented with the selected augment. The facial image 148 may be obtained from facial image data from at least one uploaded image.

[57] The processing 124 may comprise serving an image upload interface and receiving at least one uploaded image data at step 129. The i mage upload interface may prom pt the user to upload an image of a subject to be merged into the storybook.

[58] The processing 124 may com prise at least one of facial recognition, identity recognition and attribute recognition at step 130.

[59] For exam ple, for the uploaded image, the image recognition control ler 109 may perform facial recognition thereon to identify at least one facial region within the uploaded image. In em bodiments, the server 101 may com m unicate with a facial recognition service by way of an API cal l.

[60] In em bodiments, the image upload interface is configured to display an error if more than one face is detected within an uploaded image.

[61] In alternative em bodiments, the image recognition controller 109 is configured for detecting a plurality of facial regions within the uploaded images either for selecting an optimal facial image or for recognising separate subjects.

[62] For selecting optimal facial images, i n em bodiments, the merger controller 112 may be configured for selecting an optimal facial image from the pl urality of detected facial regions at step 131. For exam ple, the facial placeholder regions 149 may specify (such as by way of associated meta data) at least one of gaze orientation (such as from -90° to 90° about a vertical axis) and head tilt (such as from -45° to 45° about a frontal axis) . For exam ple, meta data associated with the facial placeholder region 149 may specify that the placeholder region 149 is for a face facing directly forward at 0°, looking directly left (-90°) or right (90°) or various degrees therebetween.

[63] As such, for the pl urality of facial regions identified by the image recognition controller 109, the merge controller 112 may select an optimal identified face optimally matching the gaze orientation specified by the facial region placeholder 149.

[64] In embodiments, the image upload interface may require the user upload sufficient images until such time that appropriate facial images are identified to satisfy the requirements of facial region placeholders 149 of the storybook tem plate. For exam ple, a storybook tem plate may com prise facial image placeholders specifying optimal gaze orientations of -90°, 0° and 90°. As such, the user may have uploaded two photographs of a child looking straight forward and left. As such, the image upload interface may prom pt the user to upload a further photograph of the child also looking to the right.

[65] In em bodiments, the image manipulation control ler 110 may dynam ical ly generate facial image data at appropriate gaze orientations. For exam ple, the image manipulation controller 110 may com prise a 3D facial model to which the 2D recognised facial image data is mapped. The image manipulation control ler 110 may then rotate the gaze orientation of the 3D facial model to extract 2D gaze adjusted facial image data for merging.

[66] In embodiments, the image manipulation control ler 110 may tilt facial image data to match a tilt orientation specified by the facial placeholder region 149.

[67] Alternatively, for recognising separate subjects, the image recognition controller 109 may be configured for perform ing identity recognition. For example, a storybook tem plate may com prise two characters such as a child and a grandparent character and wherein the image recognition control ler 109 is configured for perform ing image recognition to identify two separate subject identities from the uploaded image data corresponding to each character. The image recognition controller 109 may automatically perform separate subject identity recognition according to the facial placeholder regions 149 specified within each storybook tem plate. For exam ple, the facial placeholder regions 149 within a selected storybook tem plate may com prise meta data specifying two different cha racter types such that the image recognition controller automatical ly seeks to identify two separate subject s from the uploaded images. As such, for exam ple, for a selected storybook tem plate com prising the child and the grandparent characters, the user may upload various photographs of a child and grand parent wherein the image recognition control ler 109 automatically identifies the separate subjects therein . [68] As al luded to above, the image recognition controller 109 may also perform attribute recognition, such as gender, age category attribute recognition and the l ike at step 130.

[69] At step 132, the processing 124 may com prise the merge controller 112 extracting facial image data from the uploaded image data at the at least one facial region detected by the image recognition controller 109 and, in em bodiments perform adjustments thereon using the image manipu lation controller 110.

[70] The adjustments may com prise the above-described rotation and/or tilt adjustment.

[71] Further adjustments may be performed such as boundary cro pping wherein the image manipulation controller 110 identifies a facial boundary and extracts pixels fall ing within the facial boundary. The facial boundary may be identified by at least one of tonal and intensity edge detection.

[72] The image manipulation controller 110 may identify a hair region boundary and excl ude pixels fall ing within the hair region boundary.

[73] The image manipulation controller 110 may further perform tonal adjustments to, for exam ple, normalise skin tone. In em bodiments, the image manipulation control ler 110 may dynam ically determine at least one of tone and intensity of the image data 141 of the page tem plate 114 and adjust at least one of the tone and intensity of the facial image data accordingly to match the at least one of the tone and intensity of the image data 141 of the page tem plate 114.

[74] In alternative em bodiments, the merge control ler 112 may adjust a character avatar according to subject characteristics and generate avatar image data and merge avatar derived image data to the facial image placeholder regions 149. In em bodiments, the subject characteristics may be identified from the uploaded image data. For exam ple, the image recognition controller 109 m ay recognise that a subject within an uploaded image has brown hair and bl ue eyes. Furthermore, the user may have input the gender of the subject or alternatively the image recognition controller 109 may have automatically identified the gender of the su bject. As such, the image manipulation controller 110 may adjust a male avatar with brown hair and bl ue eyes and generate avatar image data for merging.

[75] Step 133 may com prise the image augmentation controller 111 identifying a character region and/or poise type within the page tem plate 114. Specifically, with reference to Figure 3, the page tem plate 141 may encode a character region 150. The character region 150 may further specify a poise type, such as standing, running, jum ping, lying down, sitting a nd the like poises. [76] For the selected character type 143, the merge controller 112 may merge character type image data at the character region 150 at step 134. For exam ple, for the selected boy type character 144, and for the character region 151 specifying the ju m ping poise (such as by way of associated meta data), the merge controller 112 may use boy type jum ping character image data.

[77] As al luded to above, the page tem plate 141 may com prise a pl ural ity of layers. As such, the rabbit character 153 shown in Figure 3 may be presented on a foreground layer with the surrounding forest 154 shown on a background layer. The character image data 155 may be merged by the merge control ler 112 on an intermediate layer between the foreground and background layers so as to blend between foreground and background layers 153, 154 as shown a Figure 3.

[78] The merge controller 112 may identify facial placeholder regions 149 at step 135 and merge the adjusted facial image data 156 as shown in Figure 6 at the facial placeholder regions 149 at step 136.

[79] The merge controller 112 may further util ise the image augmentation controller 112 to apply various image augments. For exam ple, the image augmentation controller 112 may augment the adjusted facial image data with the selected hat augmentati on 147 as il lustrated in Figure 5.

[80] In embodiments, as opposed to the merge controller 112 merging character and facial image data separately at separate respective locations 150, 149, the merge control ler 112 may util ise character types 143 com prising faci a l placeholder regions 149 as substantial ly shown in Figure 4. As such, in accordance with this em bodiment, the page tem plates 114 need only be encoded with character placeholders 150 which are inserted with the adjusted facial image data al ready merge the rewith.

[81] Step 137 may com prise the merge control ler 112 identifying the information placeholder regions 142 at step 137 and merging the information data field val ues therein at step 138. Specifically, Figure 6 shows the aforedescribed welcome message data f ield placeholder region 142 having the associated data value 146 therein .

[82] At step 139, the com pilation controller 113 generates com piled storybook data com prising the generated pages 117 in accordance with the page tem plates 114, facial images 116 and other information.

[83] The com piled storybook data may be made available to download an electronic e - book format. Alternatively, the com pile storybook data may be transm itted to a web print service for hardcopy printing thereof. [84] The server 101 may serve and authoring interface for storybook tem plate authoring. The interface may al low for drag-and-drop functionality to specify locations of characters, facial regions, information fields and the like in the configuration of the associated meta data. The interface may al low for the configuration of various characters 143, character poises and image augments 146.

[85] In em bodiments, page tem plate meta data may specify the placeholder regions. Furthermore, meta data may further specify placeholder region properties, such as g ait orientation and the l ike. A page tem plate 114 may com prise structured markup (such as XM L data) associated with image data.

[86] In alternative em bodiments, the system 100 is configured for identifying the placeholder region from image data. For exam ple, with reference to Figure 3, a substantially ovular blank image region 157 may be detected by the merge control ler 112 for the placement of the facial image data. I n accordance with this em bodiment, the merge controller 112 may be configured for cropping a nd/or sizing the facial image data 156 to fit within the image region 157.

[87] The foregoing description, for purposes of expla nation, used specific nomenclature to provide a thorough u nderstanding of the invention. Flowever, it will be apparent to one skil led in the art that specific details are not required in order to practi se the invention. Thus, the foregoing descriptions of specific em bodiments of the invention are presented for purposes of ill ustration and description. They are not intended to be exhaustive or to l im it the invention to the precise forms disclosed as obviously many modifications and variations are possible in view of the above teachings. The em bodiments were chosen and described in order to best explain the principles of the invention and its practical applications, thereby enabling others skilled in the art to best util ize the invention and various em bodiments with various modifications as are suited to the particular use contem plated. It is intended that the fol lowing claims and their equival ents define the scope of the invention.

[88] The term "approximately" or sim ilar as used herein should be construed as being within 10% of the value stated unless otherwise indicated.

Claims

Cla i ms

1. A storybook com pilation system com prising:

a database comprising a plural ity of storybook templates, each comprising a plurality of page templates, wherein at least one page tem plate has page image data and defines at least one placeholder region;

a storybook selection interface for the selection of a selected storybook template; an image upload interface for uploading image data com prising an image including a face; and

a merge controller which merges facial image data derived from the image data with the page image data at the at least one placeholder region for generating compiled storybook data.

2. A system as claimed in claim 1, wherein the system serves an information data field interface comprising a pl urality of data fields for receipt of a ssociated data field values and wherein the merge controller is configured for merging the data field values with the page image data.

3. A system as claimed in claim 2, wherein at least one page template define at least one data field placeholder region and wherein the system is configured for dynamically configuring the information data field interface with input data fields according to meta data associated with the at least one data field placeholder region.

4. A system as claimed in claim 1, wherein the at least one page template defines a plural ity of layers and wherein the merge controller is configured for merging the facial data at one of the layers.

5. A system as claimed in claim 1, wherein the system is configured for serving a character type selection interface for the receipt of a selected character type and wherein the merge controller is configured for merging character type image data with the page image data according to the selected character type with the page image data.

6. A system as claimed in claim 5, further comprising an image recognition controller and wherein the image recognition control ler is configured for identifying at least one attribute from the uploaded image data and for selecting a character type accordingly.

7. A system as claimed in claim 5, wherein the at least one placeholder region has associated meta data defining a character poise and wherein the merge controller is configured for selecting the character type image data according to the defined character poise.

8. A system as claimed in claim 5, wherein the system is configured for serving a character augment selection interface for the receipt of a selected character augment and wherein the merge controller is configured for merging the character type image data with augment image data according to the selected character augment.

9. A system as claimed in claim 1, further comprising an image recognition controller which performs facial image recognition on the uploaded image data to identify a facial region within the uploaded image data.

10. A system as claimed in claim 9, wherein the merge controller is configured for extracting the facial image data from the uploaded image data at the identified facial region.

11. A system as claimed in claim 10, wherein the image recognition controller is configured for identifying a facial boundary and wherein the merge controller is configured for extracting pixels from the uploaded image data within the facial boundary.

12. A system as claimed in claim 11, wherein the image recognition controller is configured for identifying a hairline boundary and wherein the merge controller is configured for excluding pixels from the uploaded image data within the hairline boundary.

13. A system as claimed in claim 10, wherein the merge control ler is configured for performing at least one of tonal and intensity adjustments on pixels extracted from the uploaded image data.

14. A system as claimed in claim 13, wherein the at least one of the total and intensity adjustments are performed in accordance with at least one of a respective tonal and intensity level of image data of the page template.

15. A system as claimed in claim 1, further comprising an image recognition controller configured to identify at least facial attribute from the uploaded image data comprising at least one of hair colour and eye colour and wherein the merge control ler is configured for augmenting a character avatar according to the at least one facial attribute and facial image data is derived from the augmented character avatar.

16. A system as claimed in claim 9, wherein the image comprises a plural ity of faces and wherein the merge controller is configured for selecting an optimal facial region therefrom.

17. A system as claimed in claim 16, wherein the at least one placeholder region has metadata defining a gaze orientation and wherein the merge controller is configured for selecting the optimal facial region in accordance with the defined gaze orientation.

18. A system as claimed in claim 17, wherein the image upload interface is configured for indicating that the uploaded image data comprises insufficient faces to satisfy defined gaze orientations of the at least one placeholder region meta data.

19. A system as claimed in claim 9, wherein the at least one placeholder region defines a gaze orientation and wherein the merge controller is configured to map two -dimensional facial image data extracted from the uploaded image data onto a 3D facial model, rotate the 3D model according to the gaze orientatio n and extract 2D gaze adjusted facial image data therefrom for merging.

20. A system as claimed in claim 9, wherein the page tem plates defines at least two placeholder regions having associated meta data defining at least two different character types and wherein the image recognition controller is further configured for identity recognition to identify faces of at least two separate subjects from the uploaded image data and wherein the merge controller merges separate facial image data extracted from the uploaded image data for each of the at least two subjects at each of the at least two respective placeholder regions.