Digital video generation method and device based on key frame

Abstract

The invention relates to the technical field of artificial intelligence, in particular to a digital video generation method and device based on key frames, equipment, a medium and a program product. The method comprises the following steps: acquiring a to-be-processed video, and segmenting the to-be-processed video into video material segments; determining a plurality of pieces of key action information in response to an application demand of the target video; determining a key frame group matched with each piece of key action information in the video material segment based on the key action information; based on the application demand, randomly sorting the video material segments, and inserting a target key frame into the sorted material segment sequence to obtain a to-be-combined material sequence; and generating an intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video. By adopting the method, the smoothness of the digital video can be improved, and the splicing trace of the video is weakened.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a key-frame-based digital video generation method, device, equipment, medium and program product. Background Art

[0002] Digital customer service is a virtual human that uses artificial intelligence technology to mimic real customer service representatives and interact with users through text, voice or images. They can understand users' questions, provide answers, and handle transactions, thereby improving customer service efficiency and user experience. Digital human technology combines multiple fields such as computer graphics, artificial intelligence, and animation technology to create virtual characters that are almost indistinguishable from real humans. Digital human video refers to video content containing digital human characters, which may be completely virtual or generated through motion capture and expression capture technology of human actors. Digital human videos usually have very high visual quality and realism, and can present vivid character images and movements. The application fields of digital human video technology are very wide, including but not limited to film and TV production, game development, virtual reality and augmented reality, advertising and marketing. In general, digital human video refers to video content containing digital human characters generated by digital human technology, with high realism and visual effects, and is widely used in movies, games, virtual reality and other fields.

[0003] In the relevant technologies, digital customer service virtual people usually have the following characteristics: intelligent interaction: using natural language processing (NLP), machine learning and other technologies to understand user intentions and make corresponding responses; multi-channel support: can provide services on multiple channels such as websites, apps, social media, etc.; 7x24 hours online: not limited by time, respond to user needs at any time; personalized service: can provide customized services based on user portraits; cost-effectiveness: reduce labor costs and improve service efficiency. In short, digital customer service virtual people are an important tool for enterprises to optimize customer service and improve operational efficiency. In the synthesis of digital videos based on digital people, the steps usually include: background production: create or select the background of the video, which can be a static picture, dynamic video or 3D scene; synthesis: synthesize the digital human model with the background. Video editing software (such as AdobeAfter Effects, Premiere Pro, etc.) can be used for synthesis; editing: edit, color, add special effects, etc. to the video to make it more attractive.

[0004] However, the current processing methods for digital video generation have the following technical problems: The video obtained through manual processing has serious synthesis traces and needs to be optimized. Summary of the invention

[0005] Based on this, it is necessary to provide a key-frame-based digital video generation method, device, computer equipment, computer-readable storage medium and computer program product that can improve the smoothness of digital video and weaken the splicing traces of the video in response to the above technical problems.

[0006] In a first aspect, the present application provides a key frame-based digital video generation method. The method comprises: Acquire a video to be processed, determine a segmentation point according to deviation comparison data between video frames of the video to be processed, and segment the video to be processed into video material segments based on the segmentation point; In response to application requirements of the target video, determining a number of key action information, wherein the key action information is used to characterize a specific digital human action; Determining a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames; Based on the application requirements, the video material segments are randomly sorted, and target key frames are inserted into the sorted material segment sequence to obtain a material sequence to be combined, wherein the target key frames are randomly selected from the corresponding key frame group; An intermediate frame between the video material segment and the key frame is generated based on a preset video generation algorithm to obtain the target video.

[0007] In one embodiment, the determining of a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames including: Acquire a reference action sequence corresponding to the key action information, wherein the reference action sequence includes a plurality of reference frames, and the reference frames include a simplified joint model of a digital human performing a specific action; A plurality of sampling points are set in the simplified joint model of the digital human, and the key frame group is determined based on deviation information of the sampling points between the reference frame and the video frame.

[0008] In one embodiment, the obtaining of a reference action sequence corresponding to the key action information, wherein the reference action sequence includes a plurality of reference frames, and the reference frames include a simplified joint model of a digital human performing a specific action, including: Determine a key action classification based on the key action information, wherein the key action classification includes a body action type and / or a facial action type; The model architecture of the digital human joint simplified model is determined based on the key action classification, wherein the model architecture includes a torso joint model and a facial capture model, and the digital human joint simplified model includes a single architecture or a combination of multiple architectures.

[0009] In one embodiment, the generating the intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video includes: Determine, based on a preset sampling window, a transitional video frame group at the head and tail of the video material segment to be connected, wherein the number of frames of the transitional video frame group is greater than 1, and the duration of the transitional video frame group is not less than a preset proportion of the total duration of the video material segment to which it belongs; An intermediate frame group is generated based on the transition video frame group and the target key frame.

[0010] In one embodiment, the generating the intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video includes: The intermediate frame group includes a plurality of intermediate frames to be inserted, and the intermediate frames are inserted into the transition video frame group and the target key frame to connect the video material segment with the key frame. The distribution form of the intermediate frames is sparse on both sides and dense in the middle.

[0011] In one embodiment, after the intermediate frame between the video material segment and the key frame is generated based on a preset video generation algorithm to obtain the target video, the method further includes: Generate a control accessory associated with the target video, and adjust the target video based on the control accessory, wherein the control accessory includes an easing function and / or a spline curve, and the spline curve includes a control point sequence having a mapping relationship with a video frame group in the target video.

[0012] In a second aspect, the present application also provides a key frame-based digital video generation device. The device comprises: A video segmentation module is used to obtain a video to be processed, determine a segmentation point according to deviation comparison data between video frames of the video to be processed, and segment the video to be processed into video material segments based on the segmentation point; A demand data module, used to determine a number of key action information in response to the application requirements of the target video, wherein the key action information is used to characterize a specific digital human action; A key frame module, configured to determine a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames; A material sequence module, used for randomly sorting the video material segments based on the application requirements, and inserting target key frames into the sorted material segment sequence to obtain a material sequence to be combined, wherein the target key frames are randomly selected from the corresponding key frame group; The video generation module is used to generate an intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video.

[0013] In one embodiment, the key frame module includes: A simplified model module, used to obtain a reference action sequence corresponding to the key action information, wherein the reference action sequence includes a plurality of reference frames, and the reference frames include a simplified joint model of a digital human performing a specific action; The sampling module is used to set a number of sampling points in the simplified joint model of the digital human, and determine the key frame group based on the deviation information of the sampling points between the reference frame and the video frame.

[0014] In one embodiment, the simplified model module includes: An action classification module, used to determine a key action classification based on the key action information, wherein the key action classification includes a body action type and / or a facial action type; A model architecture module is used to determine the model architecture of the digital human joint simplified model based on the key action classification, wherein the model architecture includes a torso joint model and a facial capture model, and the digital human joint simplified model includes a single architecture or a combination of multiple architectures.

[0015] In one embodiment, the video generation module includes: A transition frame group module, used to determine the transition video frame group at the head and tail of the video material segment to be connected based on a preset sampling window, the number of frames of the transition video frame group is greater than 1, and the duration of the transition video frame group is not less than a preset proportion of the total duration of the video material segment to which it belongs; The intermediate frame group module is used to generate an intermediate frame group based on the corresponding transition video frame group and the target key frame.

[0016] In one embodiment, the video generation module includes: The interpolation module is used for the intermediate frame group including a plurality of intermediate frames to be inserted, inserting the intermediate frames into the transition video frame group and the target key frame to connect the video material segment with the key frame, and the distribution form of the intermediate frames is sparse on both sides and dense in the middle.

[0017] In one embodiment, after the video generation module, it also includes: A control accessory module is used to generate a control accessory associated with the target video and adjust the target video based on the control accessory. The control accessory includes an easing function and / or a spline curve, and the spline curve includes a control point sequence that has a mapping relationship with a video frame group in the target video.

[0018] In a third aspect, the present application further provides a computer device, which includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps in the key-frame-based digital video generation method described in any one of the embodiments in the first aspect are implemented.

[0019] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in a key-frame-based digital video generation method as described in any one of the embodiments in the first aspect.

[0020] In a fifth aspect, the present application further provides a computer program product, which includes a computer program, and when the computer program is executed by a processor, the steps in the key-frame-based digital video generation method described in any one of the embodiments in the first aspect are implemented.

[0021] The above-mentioned key-frame-based digital video generation method, device, computer equipment, storage medium and computer program product can achieve the following beneficial effects corresponding to the technical problems in the background technology by deducing the technical features in the claims: The present application provides a keyframe-based digital video generation method, including: obtaining a video to be processed, determining a segmentation point according to the deviation comparison data between the video frames of the video to be processed, and segmenting the video to be processed into video material segments based on the segmentation point; in response to the application requirements of the target video, determining a number of key action information, the key action information is used to characterize a specific digital human action; determining a key frame group matching each key action information in the video material segment based on the key action information, the key frame group including a number of key frames; based on the application requirements, randomly sorting the video material segments, and inserting a target key frame into the sorted material segment sequence to obtain a material sequence to be combined, the target key frame is randomly selected from the corresponding key frame group; generating an intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video. In implementation, after obtaining the video to be processed, at least one cutting point is obtained through the deviation comparison data, so that the long video to be processed is cut into multiple video material segments, so that the similarity of the connection between the cut material video segments is high. Finally, by generating transition materials between the material video segments and splicing the material video segments through the intermediate frame group, it is helpful to further improve the smoothness of the connection between the material video segments, reduce the editing traces of the target video, and improve the quality of the produced digital video. On the other hand, since the key frames representing specific actions are set, if the same key frames are repeated repeatedly, the repetition of the specific actions will be high and the artificial traces will be obvious. The present application sets a key frame group containing several key frames, and improves the randomness of the key frame setting by random selection, which can eventually reduce the repetition of the key frames of the target video, improve the naturalness of the generated target video, and reduce the splicing and synthesis traces. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 A first flow chart of a key-frame-based digital video generation method in one embodiment; Figure 2 is a schematic diagram of sampling points in an embodiment; Figure 3 is a structural block diagram of a video generating device in one embodiment; Figure 4 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] In the relevant technologies, digital customer service virtual people usually have the following characteristics: intelligent interaction: using natural language processing (NLP), machine learning and other technologies to understand user intentions and make corresponding responses; multi-channel support: can provide services on multiple channels such as websites, apps, social media, etc.; 7x24 hours online: not limited by time, respond to user needs at any time; personalized service: can provide customized services based on user portraits; cost-effectiveness: reduce labor costs and improve service efficiency. In short, digital customer service virtual people are an important tool for enterprises to optimize customer service and improve operational efficiency. In the synthesis of digital videos based on digital people, the steps usually include: background production: create or select the background of the video, which can be a static picture, dynamic video or 3D scene; synthesis: synthesize the digital human model with the background. Video editing software (such as AdobeAfter Effects, Premiere Pro, etc.) can be used for synthesis; editing: edit, color, add special effects, etc. to the video to make it more attractive.

[0025] However, the current processing methods for digital video generation have the following technical problems: The video obtained through manual processing has serious synthesis traces and needs to be optimized.

[0026] Based on this, the present application provides a method for generating digital video based on key frames. In one embodiment, Figure 1 As shown, a method for generating digital video based on key frames is provided. This embodiment takes the method applied to a terminal as an example. It can be understood that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and is implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps: Step 102: obtaining a video to be processed, determining segmentation points according to deviation comparison data between video frames of the video to be processed, and segmenting the video to be processed into video material segments based on the segmentation points.

[0027] A video frame is the basic unit in video data processing, representing a static image of a video at a specific time point. Video frames are arranged in sequence in a video file or video stream to present continuous dynamic images.

[0028] The cutting point may refer to the segmentation position selected for segmenting the video to be processed, and the segmentation point may be determined based on the video time or the video frame. In the fine segmentation process, the segmentation point is usually determined based on the video frame. For example, if the 30th frame of the video to be processed is used as the segmentation point, the video before 30 frames is segmented into the front video segment, and the video after 31 frames is segmented into the back video segment.

[0029] The deviation comparison result may refer to a parameterized quantization result obtained by comparing the degree of deviation between two video frames.

[0030] Step 104: In response to the application requirements of the target video, determine a number of key action information, where the key action information is used to characterize a specific digital human action; Step 106: determining a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames; Step 108: Based on the application requirements, the video material segments are randomly sorted, and target key frames are inserted into the sorted material segment sequence to obtain a material sequence to be combined, wherein the target key frames are randomly selected from the corresponding key frame group; Step 1010: Generate an intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video.

[0031] In the above-mentioned key-frame-based digital video generation method, reasonable deduction is performed in combination with the technical features in the embodiment to achieve the following beneficial effects of solving the technical problems raised in the background technology: The present application provides a keyframe-based digital video generation method, including: obtaining a video to be processed, determining a segmentation point according to the deviation comparison data between the video frames of the video to be processed, and segmenting the video to be processed into video material segments based on the segmentation point; in response to the application requirements of the target video, determining a number of key action information, the key action information is used to characterize a specific digital human action; determining a key frame group matching each key action information in the video material segment based on the key action information, the key frame group including a number of key frames; based on the application requirements, randomly sorting the video material segments, and inserting a target key frame into the sorted material segment sequence to obtain a material sequence to be combined, the target key frame is randomly selected from the corresponding key frame group; generating an intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video. In implementation, after obtaining the video to be processed, at least one cutting point is obtained through the deviation comparison data, so that the long video to be processed is cut into multiple video material segments, so that the similarity of the connection between the cut material video segments is high. Finally, by generating an intermediate frame group between the material video segments, it helps to further improve the smoothness of the connection between the material video segments, reduce the editing traces of the target video, and improve the quality of the produced digital video. On the other hand, since the key frames representing specific actions are set, if the same key frames are repeated repeatedly, the repetition of the specific actions will be high and the artificial traces will be obvious. The present application sets a key frame group containing several key frames, and improves the randomness of the key frame setting by random selection, which can eventually reduce the repetition of the key frames of the target video, improve the naturalness of the generated target video, and reduce the splicing and synthesis traces.

[0032] In one embodiment, if Figure 2 As shown, step 108 includes: Step 202: Acquire a reference action sequence corresponding to the key action information, wherein the reference action sequence includes a plurality of reference frames, and the reference frames include a simplified joint model of a digital human performing a specific action; Step 204: setting a number of sampling points in the simplified joint model of the digital human, and determining the key frame group based on deviation information of the sampling points between the reference frame and the video frame.

[0033] In this embodiment, setting a simplified joint model helps to simplify the description difficulty of key actions and improve the recognition efficiency and accuracy of key frames.

[0034] In one embodiment, the step 202 includes: Step 402: determining a key action classification based on the key action information, wherein the key action classification includes a body action type and / or a facial action type; Step 404: Determine the model architecture of the digital human joint simplified model based on the key action classification, wherein the model architecture includes a torso joint model and a facial capture model, and the digital human joint simplified model includes a single architecture or a combination of multiple architectures.

[0035] In one embodiment, step 1010 includes: Step 502: Determine the transition video frame groups at the head and tail of the video material segment to be connected based on a preset sampling window, wherein the number of frames of the transition video frame group is greater than 1, and the duration of the transition video frame group is not less than a preset proportion of the total duration of the video material segment to which it belongs.

[0036] Step 504: Generate an intermediate frame group based on the transition video frame group and the target key frame.

[0037] In this embodiment, by incorporating multiple frames at the head and tail of the video material segment into the processing range of the connection generation algorithm, it is helpful to enhance the connection effect of the intermediate frames and improve the smoothness of the obtained composite video.

[0038] In one embodiment, step 1010 includes: Step 602: the intermediate frame group includes a plurality of intermediate frames to be inserted, and the intermediate frames are inserted into the transition video frame group and the target key frame to connect the video material segment with the key frame, and the distribution pattern of the intermediate frames is sparse on both sides and dense in the middle.

[0039] In one embodiment, after step 1010, the method further includes: Step 702: Generate a control accessory associated with the target video, and adjust the target video based on the control accessory, wherein the control accessory includes an easing function and / or a spline curve, and the spline curve includes a control point sequence having a mapping relationship with a video frame group in the target video.

[0040] In this embodiment, by setting the control accessories, it is helpful to adjust the target video according to the usage requirements, thereby improving the flexibility of video generation.

[0041] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0042] Based on the same inventive concept, the embodiment of the present application also provides a key-frame-based digital video generation device for implementing the key-frame-based digital video generation method mentioned above. The implementation solution provided by the device to solve the problem is similar to the implementation solution recorded in the above method, so the specific limitations in one or more key-frame-based digital video generation device embodiments provided below can refer to the above limitations on the key-frame-based digital video generation method, which will not be repeated here.

[0043] In one embodiment, Figure 3 As shown, a key frame-based digital video generation device is provided, comprising: a video segmentation module, a demand data module, a key frame module, a material sequence module and a video generation module, wherein: A video segmentation module is used to obtain a video to be processed, determine a segmentation point according to deviation comparison data between video frames of the video to be processed, and segment the video to be processed into video material segments based on the segmentation point; A demand data module, used to determine a number of key action information in response to the application requirements of the target video, wherein the key action information is used to characterize a specific digital human action; A key frame module, configured to determine a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames; A material sequence module, used for randomly sorting the video material segments based on the application requirements, and inserting target key frames into the sorted material segment sequence to obtain a material sequence to be combined, wherein the target key frames are randomly selected from the corresponding key frame group; The video generation module is used to generate an intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video.

[0044] In one embodiment, the key frame module includes: A simplified model module, used to obtain a reference action sequence corresponding to the key action information, wherein the reference action sequence includes a plurality of reference frames, and the reference frames include a simplified joint model of a digital human performing a specific action; The sampling module is used to set a number of sampling points in the simplified joint model of the digital human, and determine the key frame group based on the deviation information of the sampling points between the reference frame and the video frame.

[0045] In one embodiment, the simplified model module includes: An action classification module, used to determine a key action classification based on the key action information, wherein the key action classification includes a body action type and / or a facial action type; A model architecture module is used to determine the model architecture of the digital human joint simplified model based on the key action classification, wherein the model architecture includes a torso joint model and a facial capture model, and the digital human joint simplified model includes a single architecture or a combination of multiple architectures.

[0046] In one embodiment, the video generation module includes: A transition frame group module, used to determine the transition video frame group at the head and tail of the video material segment to be connected based on a preset sampling window, the number of frames of the transition video frame group is greater than 1, and the duration of the transition video frame group is not less than a preset proportion of the total duration of the video material segment to which it belongs; The intermediate frame group module is used to generate an intermediate frame group based on the corresponding transition video frame group and the target key frame.

[0047] In one embodiment, the video generation module includes: The interpolation module is used for the intermediate frame group including a plurality of intermediate frames to be inserted, inserting the intermediate frames into the transition video frame group and the target key frame to connect the video material segment with the key frame, and the distribution form of the intermediate frames is sparse on both sides and dense in the middle.

[0048] In one embodiment, after the video generation module, it also includes: A control accessory module is used to generate a control accessory associated with the target video and adjust the target video based on the control accessory. The control accessory includes an easing function and / or a spline curve, and the spline curve includes a control point sequence that has a mapping relationship with a video frame group in the target video.

[0049] Each module in the above-mentioned key-frame-based digital video generation device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0050] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a digital video generation method based on key frames is realized. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0051] Those skilled in the art will understand that Figure 4 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0052] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0053] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0054] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0055] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant laws, regulations and standards of relevant countries and regions.

[0056] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0057] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0058] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A key frame-based digital video generation method, characterized in that: The method comprises: Acquire a video to be processed, determine a segmentation point according to deviation comparison data between video frames of the video to be processed, and segment the video to be processed into video material segments based on the segmentation point; In response to application requirements of the target video, determining a number of key action information, wherein the key action information is used to characterize a specific digital human action; Determining a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames; Based on the application requirements, the video material segments are randomly sorted, and target key frames are inserted into the sorted material segment sequence to obtain a material sequence to be combined, wherein the target key frames are randomly selected from the corresponding key frame group; An intermediate frame between the video material segment and the key frame is generated based on a preset video generation algorithm to obtain the target video.

2. The key-frame-based digital video generation method according to claim 1, characterized in that: The determining of a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames including: Acquire a reference action sequence corresponding to the key action information, wherein the reference action sequence includes a plurality of reference frames, and the reference frames include a simplified joint model of a digital human performing a specific action; A plurality of sampling points are set in the simplified joint model of the digital human, and the key frame group is determined based on deviation information of the sampling points between the reference frame and the video frame.

3. The key-frame-based digital video generation method according to claim 1, characterized in that: The obtaining of a reference action sequence corresponding to the key action information, wherein the reference action sequence includes a plurality of reference frames, and the reference frame includes a simplified joint model of a digital human performing a specific action, including: Determine a key action classification based on the key action information, wherein the key action classification includes a body action type and / or a facial action type; The model architecture of the digital human joint simplified model is determined based on the key action classification, wherein the model architecture includes a torso joint model and a facial capture model, and the digital human joint simplified model includes a single architecture or a combination of multiple architectures.

4. The key-frame-based digital video generation method according to claim 1, characterized in that: The generating the intermediate frame between the video material segment and the key frame based on a preset video generating algorithm to obtain the target video comprises: Determine, based on a preset sampling window, a transitional video frame group at the head and tail of the video material segment to be connected, wherein the number of frames of the transitional video frame group is greater than 1, and the duration of the transitional video frame group is not less than a preset proportion of the total duration of the video material segment to which it belongs; An intermediate frame group is generated based on the transition video frame group and the target key frame.

5. The key-frame-based digital video generation method according to claim 4, characterized in that: The generating the intermediate frame between the video material segment and the key frame based on a preset video generating algorithm to obtain the target video comprises: The intermediate frame group includes a plurality of intermediate frames to be inserted, and the intermediate frames are inserted into the transition video frame group and the target key frame to connect the video material segment with the key frame. The distribution form of the intermediate frames is sparse on both sides and dense in the middle.

6. The key-frame-based digital video generation method according to claim 1, characterized in that: After the intermediate frame between the video material segment and the key frame is generated based on a preset video generation algorithm to obtain the target video, the method further includes: Generate a control accessory associated with the target video, and adjust the target video based on the control accessory, wherein the control accessory includes an easing function and / or a spline curve, and the spline curve includes a control point sequence having a mapping relationship with a video frame group in the target video.

7. A key frame-based digital video generation device, characterized in that: The device comprises: A video segmentation module is used to obtain a video to be processed, determine a segmentation point according to deviation comparison data between video frames of the video to be processed, and segment the video to be processed into video material segments based on the segmentation point; A demand data module, used to determine a number of key action information in response to the application requirements of the target video, wherein the key action information is used to characterize a specific digital human action; A key frame module, configured to determine a key frame group matching each key action information in the video material segment based on the key action information, wherein the key frame group includes a plurality of key frames; A material sequence module, used for randomly sorting the video material segments based on the application requirements, and inserting target key frames into the sorted material segment sequence to obtain a material sequence to be combined, wherein the target key frames are randomly selected from the corresponding key frame group; The video generation module is used to generate an intermediate frame between the video material segment and the key frame based on a preset video generation algorithm to obtain the target video.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images