Video animation generation method and system based on AIGC technology

Abstract

The present invention relates to a method for generating video animation based on AI GC technology, which belongs to the field of video animation generation. The method comprises: performing multiple learning on a Hoffet neural network to obtain a Hoffet neural network after multiple learning and outputting it as an AI GC conversion model; using the AI ​​GC conversion model to intelligently generate multiple frames of video animation pictures of a target video animation based on each piece of content information corresponding to each frame of a picture book scan image, each piece of configuration data of a target children's picture book, and each piece of configuration data of a target video animation. The present invention also relates to a method for generating video animation based on AI GC technology. Through the present invention, multiple basic data can be extracted from children's picture books that are rich and colorful and have many sources, and an AI GC conversion model with a customized structure can be introduced to complete the intelligent conversion of children's picture books to video animations based on the multiple basic data, thereby effectively expanding the generation source of video animations.

Description

Technical Field

[0001] The present invention relates to the field of video animation generation, and in particular to a video animation generation method and system based on AIGC technology. Background Art

[0002] With the rapid development of science and technology, artificial intelligence technology is increasingly used in various fields. Especially in the field of education, the introduction of artificial intelligence content generation technology, namely AIGC technology, has brought unlimited possibilities for the innovative development of children's picture books. People expect to use AIGC technology to transform the colorful and diverse children's picture books into vivid and interesting video animations, providing children with a richer and more diverse reading experience. There are also some video animation generation solutions based on AIGC technology in the prior art.

[0003] For example, the Chinese invention patent publication CN117274450A proposes an artificial intelligence-based animation image generation system and method, which first obtains the text description of the animation image to be generated input by the user, then obtains the basic image features of the animation image to be generated selected by the user, then encodes the basic image features to obtain the basic image encoding feature vector, then performs animation image context semantic feature analysis on the text description of the animation image to be generated to obtain a sequence of text context semantic feature vectors of the animation image to be generated, then performs feature embedding analysis on the sequence of the basic image encoding feature vector and the text context semantic feature vector of the animation image to be generated to obtain the image embedded text description semantic features, and finally, generates the animation image based on the image embedded text description semantic features. In this way, the animation image generation can be made more efficient, flexible and innovative.

[0004] For example, the Chinese invention patent publication CN115147521A proposes a method for generating character expression animation based on artificial intelligence semantic analysis, the method comprising receiving an audio file; converting the audio file according to a preset voice conversion rule to obtain a text sequence, wherein the text sequence includes pronunciation information of each character and the start and end time of pronunciation; obtaining corresponding lip expression sequence information according to the pronunciation information and the start and end time of pronunciation; obtaining emotional information according to the text sentiment analysis result in the text sequence; synthesizing the expression library and action library obtained through the emotional information with the lip expression sequence information to obtain the character target expression animation and send it to the user end. This application solves the problem of low matching degree between traditional lip shape and expression algorithms, so as to efficiently synthesize high-quality facial expression animation, thereby meeting the large demand for virtual character expression animation in various industries.

[0005] It can be seen that although the above-mentioned existing technologies all use artificial intelligence content generation technology, namely AI GC technology, for animation production, they are either based on AI GC technology to produce animated images, or based on AI GC technology to produce character expression animations, and use limited basic data to produce relatively simple animation content. There is a lack of a production mechanism for the entire video animation, and naturally it is impossible to convert the rich and colorful children's picture books from many sources into vivid and interesting video animations to execute the generation of the entire video animation, resulting in the rich and colorful children's picture books from many sources cannot be used for the production of video animations. Summary of the invention

[0006] In order to solve the technical problems in the prior art, the present invention provides a video animation generation method and system based on AI GC technology, which can extract sufficient and comprehensive multiple basic data from rich and colorful children's picture books from numerous sources, and introduce an AI GC conversion model with a customized structure to complete the intelligent conversion of children's picture books to video animation based on multiple basic data from children's picture books, thereby improving the utilization rate of rich and colorful children's picture books from numerous sources, enhancing the conversion efficiency of video animation, and expanding the generation sources of video animation.

[0007] According to a first aspect of the present invention, a method for generating video animation based on AI GC technology is provided, the method comprising:

[0008] Scanning the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page of the target children's picture book;

[0009] Perform content analysis on each frame of the picture book scanned image to obtain a single piece of content information corresponding to the frame of the picture book scanned image, and output each piece of content information corresponding to each frame of the picture book scanned image;

[0010] Collecting various configuration data of the target children's picture book and obtaining various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames;

[0011] Performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as the AIGC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book;

[0012] The AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation;

[0013] Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair;

[0014] Among them, content analysis is performed on each frame of the picture book scanned image to obtain a single content information corresponding to the frame of the picture book scanned image, and each content information corresponding to each frame of the picture book scanned image is output, including: the single content information corresponding to the frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point.

[0015] According to a second aspect of the present invention, a video animation generation system based on AIGC technology is provided, the system comprising a memory and one or more processors, the memory storing a computer program, the computer program being configured to be executed by the one or more processors to complete the following steps:

[0016] Scanning the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page of the target children's picture book;

[0017] Perform content analysis on each frame of the picture book scanned image to obtain a single piece of content information corresponding to the frame of the picture book scanned image, and output each piece of content information corresponding to each frame of the picture book scanned image;

[0018] Collecting various configuration data of the target children's picture book and obtaining various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames;

[0019] Performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as the AIGC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book;

[0020] The AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation;

[0021] Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair;

[0022] Among them, content analysis is performed on each frame of the picture book scanned image to obtain a single content information corresponding to the frame of the picture book scanned image, and each content information corresponding to each frame of the picture book scanned image is output, including: the single content information corresponding to the frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point.

[0023] According to a third aspect of the present invention, there is provided a video animation generation system based on AI GC technology, the system comprising:

[0024] A picture book scanning mechanism, used for scanning a target children's picture book page by page to obtain picture book scanned images of each frame corresponding to each page of the target children's picture book;

[0025] A content analysis mechanism, connected to the picture book scanning mechanism, for performing content analysis on each frame of the picture book scanned image to obtain a single piece of content information corresponding to the frame of the picture book scanned image, and outputting each piece of content information corresponding to each frame of the picture book scanned image;

[0026] A configuration acquisition mechanism is configured to acquire various configuration data of the target children's picture book and to acquire various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is acquired by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AI GC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames;

[0027] A network reconstruction mechanism, for performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book;

[0028] The animation generation mechanism is connected to the content analysis mechanism, the configuration collection mechanism and the network reconstruction mechanism respectively, and is used to adopt the AI ​​GC conversion model to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book and the configuration data of the target video animation;

[0029] Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair;

[0030] Among them, content analysis is performed on each frame of the picture book scanned image to obtain a single content information corresponding to the frame of the picture book scanned image, and each content information corresponding to each frame of the picture book scanned image is output, including: the single content information corresponding to the frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point.

[0031] Compared with the prior art, the present invention has at least the following four key inventive features:

[0032] Firstly: sufficient and comprehensive basic data are selected for the AI ​​GC conversion of the target children's picture book to the corresponding target video animation, wherein the basic data include the content information corresponding to each frame of the target children's picture book scanned image, the configuration data of the target children's picture book and the configuration data of the target video animation, thereby improving the effectiveness and stability of the video animation conversion based on the AI ​​GC technology;

[0033] Secondly: Specifically, in each piece of content information corresponding to each frame of the target children's picture book scanned image, the single piece of content information corresponding to each frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, each piece of cyan component value, each piece of magenta component value, each piece of yellow component value and each piece of black component value corresponding to each constituent pixel point, and each piece of horizontal coordinate value and each piece of vertical coordinate value corresponding to each foreground constituent pixel point, each piece of configuration data of the target children's picture book is the number of pages of the target children's picture book, the length and width of the effective painting area and the scanning resolution of the picture book scanned image, and each piece of configuration data of the target video animation is the set frame rate of the target video animation, the resolution per frame and the upper limit of the number of pixels changed between two adjacent frames;

[0034] Thirdly: an AI GC conversion model with a customized structure is introduced for AI GC conversion. The structural customization of the AI ​​GC conversion model is reflected in that the AI ​​GC conversion model is a Hoffet neural network after multiple learning. The number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, thereby further improving the effectiveness and stability of video animation conversion based on AI GC technology;

[0035] Finally: In each learning execution of the Hoffitt neural network, multiple frames of video animation obtained by converting a children's picture book using AI GC technology in the past are used as output data of the Hoffitt neural network, and each piece of content information corresponding to each frame of the scanned image of the children's picture book, each piece of configuration data of the children's picture book, and each piece of configuration data of the video animation obtained by the conversion are used as input data of the Hoffitt neural network item by item to execute this learning operation, thereby ensuring the learning effect of each learning operation of the Hoffitt neural network. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:

[0037] Figure 1 It is a technical flow chart of the video animation generation method and system based on AI GC technology according to the present invention.

[0038] Figure 2 This is a flowchart of the steps of a video animation generation method based on AI GC technology according to Embodiment 1 of the present invention.

[0039] Figure 3 This is a flowchart of the steps of a video animation generation method based on AI GC technology according to Example 2 of the present invention.

[0040] Figure 4 This is a flowchart of the steps of a video animation generation method based on AI GC technology according to Example 3 of the present invention.

[0041] Figure 5 This is a flowchart of the steps of a video animation generation method based on AI GC technology according to Example 4 of the present invention.

[0042] Figure 6 It is a structural schematic diagram of a video animation generation system based on AI GC technology according to Embodiment 5 of the present invention.

[0043] Figure 7 It is a structural schematic diagram of a video animation generation system based on AI GC technology according to Example 6 of the present invention. DETAILED DESCRIPTION

[0044] like Figure 1 As shown, a technical flow chart of a video animation generation method and system based on AI GC technology according to the present invention is given.

[0045] like Figure 1 As shown, the specific technical process of the present invention is as follows:

[0046] Step 1: Targetedly select sufficient and comprehensive basic data for the AIGC conversion of target children's picture books to corresponding target video animations, thereby improving the effectiveness and stability of video animation conversion based on AIGC technology;

[0047] For example, the various basic data include various content information corresponding to each frame of the target children's picture book scanned image, various configuration data of the target children's picture book, and various configuration data of the target video animation;

[0048] Specifically, in the content information corresponding to each frame of the target children's picture book scanned image, the single content information corresponding to each frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to each constituent pixel point, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to each foreground constituent pixel point, the configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective painting area and the scanning resolution of the picture book scanned image, and the configuration data of the target video animation are the set frame rate of the target video animation, the resolution per frame and the upper limit of the number of pixels changed between two adjacent frames;

[0049] Step 2: Introduce a customized AIGC conversion model for AIGC conversion. Children's picture books with different numbers of pages correspond to AIGC conversion models with different customized structures, thereby further improving the effectiveness and stability of video animation conversion based on AIGC technology;

[0050] For example, the structural customization of the AIGC conversion model is manifested in that the AIGC conversion model is a Hoffet neural network after multiple learning, and the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book;

[0051] Specifically, in each learning performed on the Hoffet neural network, multiple frames of video animation obtained by converting a certain children's picture book using the AIGC technology in the past are used as output data of the Hoffet neural network, and each piece of content information corresponding to each frame of the picture book scanned image of the certain children's picture book, each piece of configuration data of the certain children's picture book, and each piece of configuration data of the video animation obtained by the conversion are used as input data of the Hoffet neural network item by item to perform this learning operation, thereby ensuring the learning effect of each learning operation of the Hoffet neural network;

[0052] Step 3: Using the AI ​​GC conversion model with the customized structure in step 2 to perform intelligent conversion on the target children's picture book based on the sufficient and comprehensive basic data selected in step 1, so as to generate a multi-frame video animation screen of the target video animation corresponding to the target children's picture book;

[0053] At this time, the generated target video animation satisfies the configuration data of the target video animation in step 1, including the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames;

[0054] It can be seen that through the intelligent conversion operation based on AI GC technology, each children's picture book can be automatically converted into the corresponding video animation, which effectively expands the generation source of video animation and realizes the unmanned operation of video animation generation.

[0055] The key points of the present invention are: using colorful and diverse children's picture books as the generation source of video animation to effectively expand the generation source of video actions, targeted screening of various basic data based on AI GC conversion, and design of a customized structure of the AI ​​GC conversion model that performs AI GC conversion.

[0056] Next, the video animation generation method and system based on AI GC technology of the present invention will be specifically described in the form of embodiments.

[0057] Example 1

[0058] Figure 2 This is a flowchart of the steps of a video animation generation method based on AI GC technology according to Embodiment 1 of the present invention.

[0059] like Figure 2 As shown, the video animation generation method based on AI GC technology includes the following specific steps:

[0060] Step S1001: Scan the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page of the target children's picture book;

[0061] For example, a manual operation mode may be selected to scan each page of the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page, or an automatic operation mode may be selected to perform automatic page turning and automatic scanning of each page of the target children's picture book, so as to complete page by page scanning of each page of the target children's picture book, and obtain scanned images of each frame of the picture book corresponding to each page;

[0062] Step S1002: performing content analysis on each picture book scanned image frame to obtain a single piece of content information corresponding to the picture book scanned image frame, and outputting each piece of content information corresponding to each picture book scanned image frame;

[0063] Specifically, content analysis can be performed on each picture book scanned image frame based on the image recognition model to obtain a single piece of content information corresponding to the picture book scanned image frame, and each piece of content information corresponding to each picture book scanned image frame can be output;

[0064] Step S1003: Collecting various configuration data of the target children's picture book and obtaining various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AI GC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames;

[0065] Note that in this step, the target video animation corresponding to the target children's picture book has not been actually generated. It only provides key basic data for the intelligent conversion of the target video animation based on AI GC technology by providing various configuration data of the target video animation.

[0066] Step S1004: performing multiple learning on the Hoffet neural network to obtain a Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book;

[0067] For example, the Hoffet neural network is subjected to multiple learning to obtain the Hoffet neural network after multiple learning, and is output as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, including: the number of pages of the target children's picture book is 50 pages, and the corresponding number of times the Hoffet neural network is learned is 100; the number of pages of the target children's picture book is 60 pages, and the corresponding number of times the Hoffet neural network is learned is 150; the number of pages of the target children's picture book is 70 pages, and the corresponding number of times the Hoffet neural network is learned is 200, and so on;

[0068] Step S1005: using the AI ​​GC conversion model to intelligently generate multiple frames of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation;

[0069] Specifically, the MATLAB toolbox can be selected to complete the simulation and test of the data processing process of intelligently generating multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation by using the AI ​​GC conversion model;

[0070] Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair;

[0071] Wherein, performing content analysis on each frame of picture book scanned image to obtain a single content information corresponding to the frame of picture book scanned image, and outputting each content information corresponding to each frame of picture book scanned image, including: the single content information corresponding to the frame of picture book scanned image is the maximum object type encoding data corresponding to the frame of picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point;

[0072] The single content information corresponding to the picture book scanned image frame is the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, including: the maximum object type encoding data corresponding to the picture book scanned image frame is the object type encoding data corresponding to the object that occupies the largest number of constituent pixels in the picture book scanned image frame;

[0073] For example, the type of each object in the picture book scanned image and the number of pixels occupied by each object are identified, and the coded data of the object type corresponding to the object with the largest number of pixels occupied by each object in the picture book scanned image is output;

[0074] Specifically, different object types correspond to the encoding data of the object types with different specific values, thereby providing effective basic data for the intelligent conversion of candidate digitization;

[0075] Among them, performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, including: in each learning performed on the Hoffet neural network, using multiple frames of video animation obtained by converting a certain children's picture book using AIGC technology in a certain history as output data of the Hoffet neural network, using each piece of content information corresponding to each frame of the picture book scanned image of the certain children's picture book, each piece of configuration data of the certain children's picture book, and each piece of configuration data of the video animation obtained by the conversion as input data of the Hoffet neural network item by item, and performing this learning operation;

[0076] The single content information corresponding to the picture book scanned image frame is the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, and further includes: when the color component value of each constituent pixel of the picture book scanned image frame matches the color component value of the paper of the target children's picture book, the constituent pixel is judged to be a background constituent pixel, otherwise, the constituent pixel is judged to be a foreground constituent pixel;

[0077] And wherein, when the color component value of each pixel point of the frame picture book scanned image matches the color component value of the paper of the target children's picture book, the pixel point is judged to be a background pixel point, otherwise, the pixel point is judged to be a foreground pixel point, including: when the cyan component value, magenta component value, yellow component value and black component value of the pixel point respectively fall within the cyan component value interval, magenta component value interval, yellow component value interval and black component value interval corresponding to the paper of the target children's picture book, the pixel point is judged to be a background pixel point.

[0078] Example 2

[0079] Figure 3 This is a flowchart of the steps of a video animation generation method based on AI GC technology according to Example 2 of the present invention.

[0080] like Figure 3 As shown, Figure 2Different from the embodiment in the embodiment, in the video animation generation method based on the AI ​​GC technology, after the AI ​​GC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scan image, the configuration data of the target children's picture book and the configuration data of the target video animation, that is, after step S1005, the method further includes:

[0081] Step S1006: wirelessly transmitting the multi-frame video animation images of the target video animation intelligently generated to a remote big data service network element using a wireless communication link;

[0082] For example, using a wireless communication link to wirelessly transmit the multi-frame video animation images of the target video animation intelligently generated to a remote big data service network element includes: the wireless communication link is a time division duplex communication link or a frequency division duplex communication link.

[0083] Example 3

[0084] Figure 4 This is a flowchart of the steps of a video animation generation method based on AI GC technology according to Example 3 of the present invention.

[0085] like Figure 4 As shown, Figure 2 Different from the embodiments in the embodiment, in the video animation generation method based on the AIGC technology, after the AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book and the configuration data of the target video animation, that is, after step S1005, the method further includes:

[0086] Step S1007: using a liquid crystal display screen to continuously play multiple frames of video animation pictures of the target video animation intelligently generated to complete the playing process of the target video animation;

[0087] Alternatively, an LCD display array or an LED display array may be selected to replace the liquid crystal display screen, and used to complete the playback processing of the target video animation by continuously playing multiple frames of video animation pictures that intelligently generate the target video animation.

[0088] Example 4

[0089] Figure 5 This is a flowchart of the steps of a video animation generation method based on AIGC technology according to Example 4 of the present invention.

[0090] like Figure 5 As shown, Figure 2Different from the embodiment in, in the video animation generation method based on AIGC technology, after performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning and outputting it as the AIGC conversion model, the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, that is, after step S1004, the method further includes:

[0091] Step S1008: completing the model storage of the AIGC conversion model by storing various model parameters of the AIGC conversion model;

[0092] Specifically, completing the model storage of the AIGC conversion model by storing various model parameters of the AIGC conversion model includes: selecting an MMC storage device, a TF storage device or a FLASH flash memory to perform the storage of various model parameters of the AIGC conversion model.

[0093] Next, various method embodiments of the present invention are described in detail.

[0094] In the video animation generation method based on AIGC technology according to various method embodiments of the present invention:

[0095] The single content information corresponding to the picture book scanned image frame includes the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, and further includes: each constituent pixel has a cyan component value, a magenta component value, a yellow component value and a black component value in the CMYK color space;

[0096] Wherein, when the color component value of each pixel point of the frame picture book scanned image matches the color component value of the paper of the target children's picture book, the pixel point is judged to be a background pixel point; otherwise, judging the pixel point to be a foreground pixel point also includes: when the cyan component value of the pixel point falls outside the cyan component value interval corresponding to the paper of the target children's picture book, the magenta component value of the pixel point falls outside the magenta component value interval corresponding to the paper of the target children's picture book, the yellow component value of the pixel point falls outside the yellow component value interval corresponding to the paper of the target children's picture book, or the black component value of the pixel point falls outside the black component value interval corresponding to the paper of the target children's picture book, the pixel point is judged to be a foreground pixel point;

[0097] In this way, by matching the color component value of each pixel of the picture book scanned image with the color component value interval corresponding to the known target children's picture book paper, effective identification of whether each pixel of the picture book scanned image belongs to the foreground or background can be achieved.

[0098] And in the video animation generation method based on AI GC technology according to various method embodiments of the present invention:

[0099] The AI ​​GC conversion model is used to intelligently generate a multi-frame video animation screen of the target video animation based on each piece of content information corresponding to each frame of the picture book scanned image, each piece of configuration data of the target children's picture book, and each piece of configuration data of the target video animation, including: synchronously inputting each piece of content information corresponding to each frame of the picture book scanned image, each piece of configuration data of the target children's picture book, and each piece of configuration data of the target video animation into the AI ​​GC conversion model;

[0100] Among them, the use of the AI ​​GC conversion model to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book and the configuration data of the target video animation also includes: executing the AI ​​GC conversion model to obtain the target video animation.

[0101] Example 5

[0102] Figure 6 It is a structural schematic diagram of a video animation generation system based on AI GC technology according to Embodiment 5 of the present invention.

[0103] like Figure 6 As shown, the video animation generation system based on AI GC technology includes a memory and one or more processors, the memory stores a computer program, and the computer program is configured to be executed by the one or more processors to complete the following steps:

[0104] Step S1001: Scan the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page of the target children's picture book;

[0105] For example, a manual operation mode may be selected to scan each page of the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page, or an automatic operation mode may be selected to perform automatic page turning and automatic scanning of each page of the target children's picture book, so as to complete page by page scanning of each page of the target children's picture book, and obtain scanned images of each frame of the picture book corresponding to each page;

[0106] Step S1002: performing content analysis on each picture book scanned image frame to obtain a single piece of content information corresponding to the picture book scanned image frame, and outputting each piece of content information corresponding to each picture book scanned image frame;

[0107] Specifically, content analysis can be performed on each picture book scanned image frame based on the image recognition model to obtain a single piece of content information corresponding to the picture book scanned image frame, and each piece of content information corresponding to each picture book scanned image frame can be output;

[0108] Step S1003: Collecting various configuration data of the target children's picture book and obtaining various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AI GC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames;

[0109] Note that in this step, the target video animation corresponding to the target children's picture book has not been actually generated. It only provides key basic data for the intelligent conversion of the target video animation based on AI GC technology by providing various configuration data of the target video animation.

[0110] Step S1004: performing multiple learning on the Hoffet neural network to obtain a Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book;

[0111] For example, the Hoffet neural network is subjected to multiple learning to obtain the Hoffet neural network after multiple learning, and is output as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, including: the number of pages of the target children's picture book is 50 pages, and the corresponding number of times the Hoffet neural network is learned is 100; the number of pages of the target children's picture book is 60 pages, and the corresponding number of times the Hoffet neural network is learned is 150; the number of pages of the target children's picture book is 70 pages, and the corresponding number of times the Hoffet neural network is learned is 200, and so on;

[0112] Step S1005: using the AI ​​GC conversion model to intelligently generate multiple frames of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation;

[0113] Specifically, the MATLAB toolbox can be selected to complete the simulation and test of the data processing process of intelligently generating multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation by using the AI ​​GC conversion model;

[0114] Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair;

[0115] Wherein, performing content analysis on each frame of picture book scanned image to obtain a single content information corresponding to the frame of picture book scanned image, and outputting each content information corresponding to each frame of picture book scanned image, including: the single content information corresponding to the frame of picture book scanned image is the maximum object type encoding data corresponding to the frame of picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point;

[0116] The single content information corresponding to the picture book scanned image frame is the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, including: the maximum object type encoding data corresponding to the picture book scanned image frame is the object type encoding data corresponding to the object that occupies the largest number of constituent pixels in the picture book scanned image frame;

[0117] For example, the type of each object in the picture book scanned image and the number of pixels occupied by each object are identified, and the coded data of the object type corresponding to the object with the largest number of pixels occupied by each object in the picture book scanned image is output;

[0118] Specifically, different object types correspond to the encoding data of the object types with different specific values, thereby providing effective basic data for the intelligent conversion of candidate digitization;

[0119] Among them, performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, including: in each learning performed on the Hoffet neural network, using multiple frames of video animation obtained by converting a certain children's picture book using AIGC technology in a certain history as output data of the Hoffet neural network, using each piece of content information corresponding to each frame of the picture book scanned image of the certain children's picture book, each piece of configuration data of the certain children's picture book, and each piece of configuration data of the video animation obtained by the conversion as input data of the Hoffet neural network item by item, and performing this learning operation;

[0120] The single content information corresponding to the picture book scanned image frame is the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, and further includes: when the color component value of each constituent pixel of the picture book scanned image frame matches the color component value of the paper of the target children's picture book, the constituent pixel is judged to be a background constituent pixel, otherwise, the constituent pixel is judged to be a foreground constituent pixel;

[0121] And wherein, when the color component value of each pixel point of the frame picture book scanned image matches the color component value of the paper of the target children's picture book, the pixel point is judged to be a background pixel point, otherwise, judging the pixel point to be a foreground pixel point includes: when the cyan component value, magenta component value, yellow component value and black component value of the pixel point respectively fall within the cyan component value interval, magenta component value interval, yellow component value interval and black component value interval corresponding to the paper of the target children's picture book, the pixel point is judged to be a background pixel point;

[0122] like Figure 6 As shown, N processors are provided, exemplarily, where N is a natural number greater than or equal to 1.

[0123] Example 6

[0124] Figure 7 It is a structural schematic diagram of a video animation generation system based on AI GC technology according to Example 6 of the present invention.

[0125] like Figure 7 As shown, the video animation generation system based on AI GC technology includes the following components:

[0126] A picture book scanning mechanism, used for scanning a target children's picture book page by page to obtain picture book scanned images of each frame corresponding to each page of the target children's picture book;

[0127] For example, a manual operation mode may be selected to scan each page of the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page, or an automatic operation mode may be selected to perform automatic page turning and automatic scanning of each page of the target children's picture book, so as to complete page by page scanning of each page of the target children's picture book, and obtain scanned images of each frame of the picture book corresponding to each page;

[0128] A content analysis mechanism, connected to the picture book scanning mechanism, for performing content analysis on each frame of the picture book scanned image to obtain a single piece of content information corresponding to the frame of the picture book scanned image, and outputting each piece of content information corresponding to each frame of the picture book scanned image;

[0129] Specifically, content analysis can be performed on each picture book scanned image frame based on the image recognition model to obtain a single piece of content information corresponding to the picture book scanned image frame, and each piece of content information corresponding to each picture book scanned image frame can be output;

[0130] A configuration acquisition mechanism is configured to acquire various configuration data of the target children's picture book and to acquire various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is acquired by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AI GC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames;

[0131] Note that in this step, the target video animation corresponding to the target children's picture book has not been actually generated. It only provides key basic data for the intelligent conversion of the target video animation based on AI GC technology by providing various configuration data of the target video animation.

[0132] A network reconstruction mechanism, for performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book;

[0133] For example, the Hoffet neural network is subjected to multiple learning to obtain the Hoffet neural network after multiple learning, and is output as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, including: the number of pages of the target children's picture book is 50 pages, and the corresponding number of times the Hoffet neural network is learned is 100; the number of pages of the target children's picture book is 60 pages, and the corresponding number of times the Hoffet neural network is learned is 150; the number of pages of the target children's picture book is 70 pages, and the corresponding number of times the Hoffet neural network is learned is 200, and so on;

[0134] The animation generation mechanism is connected to the content analysis mechanism, the configuration collection mechanism and the network reconstruction mechanism respectively, and is used to adopt the AI ​​GC conversion model to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book and the configuration data of the target video animation;

[0135] Specifically, the MATLAB toolbox can be selected to complete the simulation and test of the data processing process of intelligently generating multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation by using the AI ​​GC conversion model;

[0136] Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair;

[0137] Wherein, performing content analysis on each frame of picture book scanned image to obtain a single content information corresponding to the frame of picture book scanned image, and outputting each content information corresponding to each frame of picture book scanned image, including: the single content information corresponding to the frame of picture book scanned image is the maximum object type encoding data corresponding to the frame of picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point;

[0138] The single content information corresponding to the picture book scanned image frame is the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, including: the maximum object type encoding data corresponding to the picture book scanned image frame is the object type encoding data corresponding to the object that occupies the largest number of constituent pixels in the picture book scanned image frame;

[0139] For example, the type of each object in the picture book scanned image and the number of pixels occupied by each object are identified, and the coded data of the object type corresponding to the object with the largest number of pixels occupied by each object in the picture book scanned image is output;

[0140] Specifically, different object types correspond to the encoding data of the object types with different specific values, thereby providing effective basic data for the intelligent conversion of candidate digitization;

[0141] Among them, performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, including: in each learning performed on the Hoffet neural network, using multiple frames of video animation obtained by converting a certain children's picture book using AIGC technology in a certain history as output data of the Hoffet neural network, using each piece of content information corresponding to each frame of the picture book scanned image of the certain children's picture book, each piece of configuration data of the certain children's picture book, and each piece of configuration data of the video animation obtained by the conversion as input data of the Hoffet neural network item by item, and performing this learning operation;

[0142] The single content information corresponding to the picture book scanned image frame is the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, and further includes: when the color component value of each constituent pixel of the picture book scanned image frame matches the color component value of the paper of the target children's picture book, the constituent pixel is judged to be a background constituent pixel, otherwise, the constituent pixel is judged to be a foreground constituent pixel;

[0143] And wherein, when the color component value of each pixel point of the frame picture book scanned image matches the color component value of the paper of the target children's picture book, the pixel point is judged to be a background pixel point, otherwise, the pixel point is judged to be a foreground pixel point, including: when the cyan component value, magenta component value, yellow component value and black component value of the pixel point respectively fall within the cyan component value interval, magenta component value interval, yellow component value interval and black component value interval corresponding to the paper of the target children's picture book, the pixel point is judged to be a background pixel point.

[0144] In addition, the present invention may also cite the following technical contents to highlight the significant technical progress of the present invention:

[0145] Performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, including: using a numerical mapping formula to represent the numerical mapping relationship of the number of times the Hoffet neural network is learned and the number of pages of the target children's picture book;

[0146] Among them, performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AI GC conversion model, the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, and further includes: in the numerical mapping formula, the number of times the Hoffet neural network is learned is used as the output parameter of the numerical mapping formula, and the number of pages of the target children's picture book is used as the input parameter of the numerical mapping formula;

[0147] For example, a numerical simulation mode is selected to perform multiple learning of the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and the simulation and test of the data processing process as the output of the AI ​​GC conversion model;

[0148] Among them, the single content information corresponding to the frame of picture book scanned image is the maximum object type encoding data corresponding to the frame of picture book scanned image, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to each pixel point, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to each foreground pixel point, including: the value range of the cyan component value, magenta component value, yellow component value or black component value of each pixel point is between 0-255.

[0149] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0150] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device / electronic device / computer-readable storage medium / computer program product embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0151] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A video animation generation method based on AIGC technology, characterized in that: The method comprises: Scanning the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page of the target children's picture book; Perform content analysis on each frame of the picture book scanned image to obtain a single piece of content information corresponding to the frame of the picture book scanned image, and output each piece of content information corresponding to each frame of the picture book scanned image; Collecting various configuration data of the target children's picture book and obtaining various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames; Performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as the AIGC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book; The AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation; Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair; Among them, content analysis is performed on each frame of the picture book scanned image to obtain a single content information corresponding to the frame of the picture book scanned image, and each content information corresponding to each frame of the picture book scanned image is output, including: the single content information corresponding to the frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point.

2. The video animation generation method based on AIGC technology as claimed in claim 1, characterized in that: The single content information corresponding to the picture book scanned image frame is the maximum object type encoding data corresponding to the picture book scanned image frame, the respective cyan component values, the respective magenta component values, the respective yellow component values ​​and the respective black component values ​​corresponding to the respective constituent pixels, and the respective horizontal coordinate values ​​and the respective vertical coordinate values ​​corresponding to the respective foreground constituent pixels, including: the maximum object type encoding data corresponding to the picture book scanned image frame is the encoding data of the object type corresponding to the object occupying the largest number of constituent pixels in the picture book scanned image frame; Among them, the Hoffitt neural network is subjected to multiple learning to obtain the Hoffitt neural network after multiple learning, and is output as an AIGC conversion model, and the number of times the Hoffitt neural network is learned is positively correlated with the number of pages of the target children's picture book, including: in each learning performed on the Hoffitt neural network, multiple frames of video animation obtained by converting a certain children's picture book using AIGC technology in history are used as output data of the Hoffitt neural network, and each piece of content information corresponding to each frame of the picture book scanned image of the certain children's picture book, each piece of configuration data of the certain children's picture book, and each piece of configuration data of the video animation obtained by the conversion are used as input data of the Hoffitt neural network item by item to perform this learning operation.

3. The video animation generation method based on AIGC technology as claimed in claim 2, characterized in that: The single content information corresponding to the frame of picture book scanned image is the maximum object type encoding data corresponding to the frame of picture book scanned image, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, and further includes: when the color component value of each constituent pixel of the frame of picture book scanned image matches the color component value of the paper of the target children's picture book, the constituent pixel is judged to be a background constituent pixel, otherwise, the constituent pixel is judged to be a foreground constituent pixel; Among them, when the color component value of each pixel point of the frame picture book scanned image matches the color component value of the paper of the target children's picture book, the pixel point is judged to be a background pixel point, otherwise, the pixel point is judged to be a foreground pixel point, including: when the cyan component value, magenta component value, yellow component value and black component value of the pixel point respectively fall within the cyan component value interval, magenta component value interval, yellow component value interval and black component value interval corresponding to the paper of the target children's picture book, the pixel point is judged to be a background pixel point.

4. The video animation generation method based on AIGC technology as claimed in claim 3, characterized in that: After the AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation, the method further includes: A wireless communication link is used to wirelessly transmit the multi-frame video animation images of the target video animation intelligently generated to a remote big data service network element.

5. The video animation generation method based on AIGC technology as claimed in claim 3, characterized in that: After the AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation, the method further includes: The target video animation is played back by a liquid crystal display screen through continuous playback of multiple frames of video animation pictures intelligently generated by the target video animation.

6. The video animation generation method based on AIGC technology as claimed in claim 3, characterized in that: After performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning and outputting it as the AIGC conversion model, and the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book, the method further includes: The model storage of the AIGC conversion model is completed by storing various model parameters of the AIGC conversion model.

7. The video animation generation method based on AIGC technology according to any one of claims 3 to 6, characterized in that: The single content information corresponding to the picture book scanned image frame includes the maximum object type encoding data corresponding to the picture book scanned image frame, the cyan component values, magenta component values, yellow component values ​​and black component values ​​corresponding to the constituent pixels, and the horizontal coordinate values ​​and vertical coordinate values ​​corresponding to the foreground constituent pixels, and further includes: each constituent pixel has a cyan component value, a magenta component value, a yellow component value and a black component value in the CMYK color space; Among them, when the color component value of each pixel point of the frame picture book scanned image matches the color component value of the paper of the target children's picture book, the pixel point is judged to be a background pixel point; otherwise, judging the pixel point to be a foreground pixel point also includes: when the cyan component value of the pixel point falls outside the cyan component value interval corresponding to the paper of the target children's picture book, the magenta component value of the pixel point falls outside the magenta component value interval corresponding to the paper of the target children's picture book, the yellow component value of the pixel point falls outside the yellow component value interval corresponding to the paper of the target children's picture book, or the black component value of the pixel point falls outside the black component value interval corresponding to the paper of the target children's picture book, the pixel point is judged to be a foreground pixel point.

8. The video animation generation method based on AIGC technology according to any one of claims 3 to 6, characterized in that: The AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation, including: synchronously inputting the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation into the AIGC conversion model; Among them, the use of the AI ​​GC conversion model to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book and the configuration data of the target video animation also includes: executing the AIGC conversion model to obtain the target video animation.

9. A video animation generation system based on AIGC technology, characterized in that: The system comprises a memory and one or more processors, the memory storing a computer program, the computer program being configured to be executed by the one or more processors to perform the following steps: Scanning the target children's picture book page by page to obtain scanned images of each frame of the picture book corresponding to each page of the target children's picture book; Perform content analysis on each frame of the picture book scanned image to obtain a single piece of content information corresponding to the frame of the picture book scanned image, and output each piece of content information corresponding to each frame of the picture book scanned image; Collecting various configuration data of the target children's picture book and obtaining various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames; Performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as the AIGC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book; The AIGC conversion model is used to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book, and the configuration data of the target video animation; Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair; Among them, content analysis is performed on each frame of the picture book scanned image to obtain a single content information corresponding to the frame of the picture book scanned image, and each content information corresponding to each frame of the picture book scanned image is output, including: the single content information corresponding to the frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point.

10. A video animation generation system based on AIGC technology, characterized in that: The system comprises: A picture book scanning mechanism, used for scanning a target children's picture book page by page to obtain picture book scanned images of each frame corresponding to each page of the target children's picture book; A content analysis mechanism, connected to the picture book scanning mechanism, for performing content analysis on each frame of the picture book scanned image to obtain a single piece of content information corresponding to the frame of the picture book scanned image, and outputting each piece of content information corresponding to each frame of the picture book scanned image; A configuration acquisition mechanism is configured to acquire various configuration data of the target children's picture book and to acquire various configuration data of the target video animation, wherein the various configuration data of the target children's picture book are the number of pages of the target children's picture book, the length and width of the effective drawing area, and the scanning resolution of the scanned image of the picture book, and the target video animation is acquired by converting the target children's picture book based on artificial intelligence content generation technology, i.e., AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of pixels changed between two adjacent frames; A network reconstruction mechanism, for performing multiple learning on the Hoffet neural network to obtain the Hoffet neural network after multiple learning, and outputting it as an AIGC conversion model, wherein the number of times the Hoffet neural network is learned is positively correlated with the number of pages of the target children's picture book; An animation generation mechanism is connected to the content analysis mechanism, the configuration collection mechanism and the network reconstruction mechanism respectively, and is used to adopt an AIGC conversion model to intelligently generate multiple frames of video animation images of the target video animation based on the content information corresponding to each frame of the picture book scanned image, the configuration data of the target children's picture book and the configuration data of the target video animation; Among them, the target video animation is obtained by converting the target children's picture book based on artificial intelligence content generation technology, namely AIGC technology, and the various configuration data of the target video animation are the set frame rate of the target video animation, the resolution of each frame, and the upper limit of the number of changed pixels between two adjacent frames, including: the target video animation is composed of a multi-frame video animation picture based on the set frame rate, and the upper limit of the number of changed pixels between two adjacent frames of the target video animation is the upper limit of the number of same-position difference pixel pairs between the previous frame of the video animation picture and the next frame of the video animation picture in the target video animation, wherein, in the target video animation, when the two pixel values ​​corresponding to the two pixel points at the same position of the previous frame of the video animation picture and the next frame of the video animation picture are different, the pixel point pair composed of the two pixel points at the same position is regarded as a same-position difference pixel point pair; Among them, content analysis is performed on each frame of the picture book scanned image to obtain a single content information corresponding to the frame of the picture book scanned image, and each content information corresponding to each frame of the picture book scanned image is output, including: the single content information corresponding to the frame of the picture book scanned image is the maximum object type encoding data corresponding to the frame of the picture book scanned image, each cyan component value, each magenta component value, each yellow component value and each black component value corresponding to each constituent pixel point, and each horizontal coordinate value and each vertical coordinate value corresponding to each foreground constituent pixel point.

Images