Video processing method and device, computer equipment and storage medium

Abstract

The invention relates to a video processing method and device, computer equipment and a storage medium. The method comprises the following steps: identifying an advertisement insertion position in a currently played video, namely identifying a to-be-implanted pixel area, and recommending preset advertisement content conforming to advertisement content preference corresponding to a target account number to the target account number according to an operation record of the target account number for historical advertisement content, and the recommended preset advertisement content is seamlessly implanted into the to-be-implanted pixel region, so that dynamic implantation and personalized recommendation of the advertisement content are realized, and the problem of how to realize personalized and targeted advertisement content delivery for different audiences is solved.

Description

Technical Field

[0001] The present application relates to the field of video processing technology, and in particular to a video processing method, apparatus, computer equipment and storage medium. Background Art

[0002] With the development of the Internet, hundreds of millions of users are active on video platforms every day, especially when watching long videos (such as TV series and movies). These users are potential advertising audiences. Advertisers usually place advertisements in popular dramas by displaying brand logos in scenes in the drama (such as billboards in the corridors of the drama). However, the traditional method of advertising placement requires reaching an agreement with the advertiser in advance during the shooting stage and placing the advertiser's brand logo in the scene on the set. This method is not only time-consuming and labor-intensive, but also lacks flexibility and cannot achieve personalized and targeted advertising content delivery for different audiences. Summary of the invention

[0003] The present application provides a video processing method, apparatus, computer device and storage medium to solve the problem of how to achieve personalized and targeted advertising content delivery for different audiences.

[0004] In a first aspect, the present application provides a video processing method, the video processing method comprising:

[0005] Obtaining the currently played video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account on the historical advertising content;

[0006] Identify and process the key frame image to be played in the currently played video to obtain a plurality of pixel regions to be implanted;

[0007] recommending a plurality of preset advertising contents to the target account according to the historical interaction data;

[0008] Implanting the preset advertising content into the pixel area to be implanted, and fusing it with the key frame image where the pixel area to be implanted is located, to obtain a key frame fused image;

[0009] When the playback frame of the currently playing video reaches a key frame, the corresponding key frame fusion image is provided to the target account.

[0010] In a second aspect, the present application provides a video processing device, the device comprising:

[0011] An acquisition module, used to acquire the currently played video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account on the historical advertising content;

[0012] A region recognition module is used to recognize and process the key frame image to be played in the currently played video to obtain a plurality of pixel regions to be implanted;

[0013] An advertisement recommendation module, configured to recommend a plurality of preset advertisement contents to the target account according to the historical interaction data;

[0014] An image fusion module, used for implanting the preset advertisement content into the pixel region to be implanted, and fusing it with the key frame image where the pixel region to be implanted is located, to obtain a key frame fusion image;

[0015] The output module is used to provide the corresponding key frame fusion image to the target account when the playback frame of the currently playing video reaches the key frame.

[0016] In a third aspect, the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned video processing method when executing the computer program.

[0017] In a fourth aspect, the present application also provides a computer storage medium storing computer executable instructions, wherein the computer executable instructions are used to execute the above-mentioned video processing method.

[0018] The above-mentioned technical scheme provided by the embodiment of the present application has the following advantages over the prior art: the method provided by the embodiment of the present application obtains the currently playing video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account for the historical advertising content; the key frame image to be played in the currently playing video is identified and processed to obtain a plurality of pixel areas to be implanted; a plurality of preset advertising contents are recommended to the target account based on the historical interaction data; the preset advertising contents are implanted into the pixel areas to be implanted, and are merged with the key frame images where the pixel areas to be implanted are located to obtain a key frame fusion image; when the playback frame of the currently playing video reaches the key frame, the corresponding key frame fusion image is provided to the target account.

[0019] Based on the above video processing method, the advertisement insertion position is identified in the currently playing video, that is, the pixel area to be implanted is identified, and according to the target account's operation record on the historical advertising content, preset advertising content that meets the target account's corresponding advertising content preference is recommended to the target account, and the recommended preset advertising content is seamlessly implanted into the pixel area to be implanted, thereby realizing dynamic implantation and personalized recommendation of advertising content, and solving the problem of how to achieve personalized and targeted advertising content delivery for different audiences. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0021] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0022] One or more embodiments are exemplarily described by pictures in the corresponding drawings, and these exemplified descriptions do not constitute limitations on the embodiments. Elements with the same reference numerals in the drawings represent similar elements, and unless otherwise stated, the figures in the drawings do not constitute proportional limitations.

[0023] Figure 1 A schematic diagram of a video processing method provided in an embodiment of the present application;

[0024] Figure 2 A structural block diagram of a video processing device provided in an embodiment of the present application;

[0025] Figure 3 A schematic diagram of the internal structure of a computer device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0027] The disclosure below provides many different embodiments or examples to implement different structures of the present invention. In order to simplify the disclosure of the present invention, the parts and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present invention. In addition, the present invention can repeat reference numbers and / or letters in different examples. This repetition is for the purpose of simplification and clarity, and does not itself indicate the relationship between the various embodiments and / or settings discussed.

[0028] In one embodiment, Figure 1 is a flow chart of a video processing method in one embodiment, referring to Figure 1, provides a video processing method. This embodiment mainly uses the method applied to a server or a terminal equipped with video software as an example to illustrate, and the video processing method specifically includes the following steps:

[0029] Step S210, obtaining the currently playing video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account on the historical advertising content.

[0030] Specifically, the target account refers to the video account of the user who logs in to the video platform to access video resources. The account type of the target account is used to correspond to different access rights of the video platform. The account types are divided into ordinary accounts, ordinary member accounts, premium member accounts, etc. Different account types are used to indicate different account levels. The higher the account level, the higher the access right level for the video platform. The higher the access right level means more video resources are authorized to access and more choices for video resource playback methods. The account level will also affect the number of key frame images in the currently playing video. For example, the higher the account level, the fewer the number of key frame images in the currently playing video, so as to reduce the number of advertisements viewed by users corresponding to premium member accounts, thereby optimizing the user viewing experience; the lower the account level, the more key frame images in the currently playing video, that is, there is also a corresponding relationship between the account level and the number of key frame images.

[0031] The currently playing video is the video resource being played by the target account, and the currently playing video includes played video segments and unplayed video segments. The played video segments and unplayed video segments are divided specifically according to the playback progress. As the playback progress changes dynamically, the unplayed video segments will be gradually divided into played video segments.

[0032] The historical interaction data indicates the target account's operation records for historical advertising content. The operation records specifically include click records, collection records, consumption records, etc., which are used to indicate the consumption behavior of the user corresponding to the target account after viewing the historical advertising content.

[0033] Step S220, identifying and processing the key frame image to be played in the currently playing video to obtain a plurality of pixel regions to be implanted.

[0034] Specifically, the key frame images to be played in the currently playing video are identified to identify the pixel areas to be implanted for inserting advertisements in each key frame image. The pixel areas to be implanted can be the full-frame pixel areas of the key frame images, that is, the pixel areas to be implanted are the entire pixel areas of the key frame images; or, the pixel areas to be implanted are local areas in the key frame images.

[0035] Step S230: recommend multiple preset advertising contents to the target account based on the historical interaction data.

[0036] Specifically, the target account's preferred advertising content can be reflected through historical interaction data, so multiple preset advertising content that the target account is interested in can be recommended to the target account based on the historical interaction data. Specifically, the historical advertising content that the target account is interested in can be determined based on the historical interaction data, and then the advertising content that is related to or has a high degree of similarity to the historical advertising content that the target account is interested in can be determined as the preset advertising content, where the historical advertising content that the target account is interested in refers to the advertising content that has been marked by the target account with click, favorite, purchase, and other operation marks.

[0037] Step S240, implanting the preset advertisement content into the pixel region to be implanted, and merging the preset advertisement content with the key frame image where the pixel region to be implanted is located, to obtain a key frame fused image.

[0038] Specifically, the preset advertising content is implanted into the pixel area to be implanted and fused with the key frame image where the pixel area to be implanted is located, that is, the preset advertising content is seamlessly implanted to obtain a key frame fusion image. The key frame fusion image is the key frame image with the advertising content changed. Since different video accounts have different preferences for advertising content, different advertising contents will be implanted in the key frame images of the same playback video for different video accounts.

[0039] Step S250, when the playback frame of the currently playing video reaches a key frame, the corresponding key frame fusion image is provided to the target account.

[0040] Specifically, when the playback frame of the currently playing video reaches a key frame, the key frame fusion image corresponding to the key frame is displayed on the client corresponding to the target account, that is, the preset advertising content of interest to the target account is personalized for the target account.

[0041] Based on the above video processing method, the advertisement insertion position is identified in the currently playing video, that is, the pixel area to be implanted is identified, and according to the target account's operation record on the historical advertising content, preset advertising content that meets the target account's corresponding advertising content preference is recommended to the target account, and the recommended preset advertising content is seamlessly implanted into the pixel area to be implanted, thereby realizing dynamic implantation and personalized recommendation of advertising content, and solving the problem of how to achieve personalized and targeted advertising content delivery for different audiences.

[0042] In one embodiment, the identifying and processing the key frame image to be played in the currently played video to obtain a plurality of pixel regions to be implanted includes:

[0043] Identify key frames to be played in the currently playing video to obtain multiple key frame images;

[0044] An instance segmentation model is used to perform instance segmentation processing on each of the key frame images, and at least one pixel region to be implanted is determined in each of the key frame images.

[0045] Specifically, the currently playing video is divided into played video segments and unplayed video segments based on the playing frames, key frames are identified in the unplayed video segments to obtain multiple key frame images, and the key frame images are subjected to instance segmentation processing using an instance segmentation model to identify and distinguish different pixel areas to be implanted in the key frame images, that is, a key frame image includes at least one pixel area to be implanted, and the pixel area to be implanted is the area in the key frame image used to display advertising content.

[0046] The instance segmentation model combines the advantages of object detection and semantic segmentation, and can perform more sophisticated and accurate image analysis. Classic instance segmentation algorithms include the image segmentation network UNet++ and the advanced semantic segmentation model DeepLabv3+ proposed by Google. The instance segmentation model not only classifies the pixels in the image, but also distinguishes different instances of the same category of objects. For example, in an image, not only all "ad slot" pixels must be identified, but also each ad slot must be distinguished. The ad slot is the pixel area to be implanted.

[0047] In one embodiment, before obtaining the currently played video and historical interaction data corresponding to the target account, the method further includes:

[0048] Get multiple key frame images with pixel area selection marks;

[0049] The preset segmentation model is trained using a plurality of key frame images with pixel region selection marks to obtain the instance segmentation model.

[0050] Specifically, the key frame image is an image manually marked as a key frame by an operator in the playing video, that is, the key frame image is determined in the playing video based on the acquired key frame marking instruction, and then the pixel area indicating the advertising position is determined in the key frame image according to the acquired framing instruction, and a pixel area framing mark is marked for the pixel area framed by the framing instruction, that is, the operator manually frames and marks the advertising position for embedding the advertisement in the key frame image using a data labeling tool (LabelMe), thereby obtaining a key frame image with a pixel area framing mark, and the key frame image with a pixel area framing mark can be called advertising position instance segmentation label data.

[0051] The preset segmentation model is trained using a plurality of key frame images with pixel region selection marks, for example, a DeepLabv3 model is trained to obtain the instance segmentation model.

[0052] In one embodiment, the recommending a plurality of preset advertising contents to the target account according to the historical interaction data includes:

[0053] Inputting the historical interaction data into an advertisement recommendation model, and determining a plurality of advertisement contents to be recommended sorted according to recommendation priorities according to output contents of the advertisement recommendation model;

[0054] A preset number of advertisement contents to be recommended among the plurality of advertisement contents to be recommended sorted according to recommendation priorities are determined as preset advertisement contents, wherein the preset number is equal to the number of the plurality of pixel areas to be implanted.

[0055] Specifically, the historical interaction data can also be directly input as input content into the advertising recommendation model, and multiple advertising contents to be recommended sorted by priority can be obtained according to the output content of the advertising recommendation model, that is, the advertising recommendation model has advertising recommendation function and recommendation sorting function. Among the multiple advertising contents to be recommended output by the advertising recommendation model, the first K advertising contents to be recommended in the recommendation priority sorting are selected as multiple preset advertising contents, that is, the preset number is K, and the preset number is also the number of pixel areas to be implanted, so as to ensure that each pixel area to be implanted has corresponding preset advertising content for display. The advertising recommendation model uses data analysis and machine learning technology to provide personalized, relevant and customized advertising content for each user. The advertising recommendation model can improve the click-through rate and user satisfaction of advertisements, thereby achieving higher conversion rate and advertising revenue. The advertising recommendation model can be implemented using the youtubeDNN algorithm model or the wide&deep algorithm model.

[0056] In one embodiment, before obtaining the currently played video and historical interaction data corresponding to the target account, the method further includes:

[0057] Acquire advertising operation data of different video accounts for playing advertisements from a database, wherein the advertising operation data has a consumption tag;

[0058] The preset recommendation model is trained according to the advertisement operation data with consumption tags to obtain the advertisement recommendation model.

[0059] Specifically, the database stores the advertising operation data of multiple different video accounts for playing advertisements. The advertising operation data refers to the operation records of the user on the advertisements embedded in the video resources during the process of watching the video resources through the video account, including click operations, favorite operations, purchase operations, etc. The advertising operation data has consumption tags, and the consumption tags specifically include click tags corresponding to click operations, favorite tags corresponding to favorite operations, and purchase tags corresponding to purchase operations, etc.

[0060] The preset recommendation model is trained using the advertising operation data with consumption labels, such as training a wide&deep model, to obtain an advertising recommendation model.

[0061] In one embodiment, the step of implanting the preset advertisement content into the pixel region to be implanted and fusing it with the key frame image where the pixel region to be implanted is located to obtain a key frame fused image includes:

[0062] The preset advertisement content replaces the current content of the pixel area to be implanted by using an image synthesis algorithm, and is merged with the key frame image where the pixel area to be implanted is located to obtain a key frame fusion image.

[0063] Specifically, the image synthesis algorithm is a technology used to seamlessly synthesize objects from different images into one image. This process usually includes detecting and extracting the target object, adjusting its size and shape, adapting it to the lighting and tone in the scene, and synthesizing it into the background image. The image synthesis algorithm can be specifically a Poisson Image Editing algorithm or a Generative Adversarial Networks (GANs) algorithm.

[0064] An image synthesis algorithm is used to replace the current content of the pixel area to be implanted with preset advertising content. That is, the current content of the pixel area to be implanted is first erased, and then the preset advertising content is added to the pixel area to be implanted. Target synthesis is performed so that the preset advertising content can be seamlessly implanted into the pixel area to be implanted, so as to obtain a key frame fusion image.

[0065] In one embodiment, the step of implanting the preset advertisement content into the pixel region to be implanted and fusing it with the key frame image where the pixel region to be implanted is located to obtain a key frame fused image includes:

[0066] The preset advertisement content is used to cover the current content of the pixel area to be implanted by using an image synthesis algorithm, and is merged with the key frame image where the pixel area to be implanted is located to obtain a key frame fused image.

[0067] Specifically, there is no need to erase the current content of the pixel area to be implanted, and the preset advertisement content is used to cover the current content of the pixel area to be implanted by using an image synthesis algorithm, and target synthesis is performed so that the preset advertisement content can be seamlessly implanted into the pixel area to be implanted, so as to obtain a key frame fusion image. This can speed up the target synthesis speed and improve the target synthesis efficiency.

[0068] Figure 1 FIG. 1 is a flow chart of a video processing method in one embodiment. It should be understood that although Figure 1The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0069] In one embodiment, Figure 2 As shown, a video processing device is provided, comprising:

[0070] The acquisition module 310 is used to acquire the currently played video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account on the historical advertising content;

[0071] The region recognition module 320 is used to recognize and process the key frame image to be played in the currently played video to obtain a plurality of pixel regions to be implanted;

[0072] An advertisement recommendation module 330, configured to recommend a plurality of preset advertisement contents to the target account according to the historical interaction data;

[0073] An image fusion module 340 is used to use the preset advertising content to be implanted in the pixel area to be implanted, and fuse it with the key frame image where the pixel area to be implanted is located to obtain a key frame fusion image;

[0074] The output module 350 is used to provide the target account with the corresponding key frame fusion image when the playback frame of the currently playing video reaches a key frame.

[0075] In one embodiment, the region identification module 320 is further configured to:

[0076] Identify key frames to be played in the currently playing video to obtain multiple key frame images;

[0077] An instance segmentation model is used to perform instance segmentation processing on each of the key frame images, and at least one pixel region to be implanted is determined in each of the key frame images.

[0078] In one embodiment, the apparatus further comprises a training module for:

[0079] Get multiple key frame images with pixel area selection marks;

[0080] The preset segmentation model is trained using a plurality of key frame images with pixel region selection marks to obtain the instance segmentation model.

[0081] In one embodiment, the advertisement recommendation module 330 is further configured to:

[0082] Inputting the historical interaction data into an advertisement recommendation model, and determining a plurality of advertisement contents to be recommended sorted according to recommendation priorities according to output contents of the advertisement recommendation model;

[0083] A preset number of advertisement contents to be recommended among the plurality of advertisement contents to be recommended sorted according to recommendation priorities are determined as preset advertisement contents, wherein the preset number is equal to the number of the plurality of pixel areas to be implanted.

[0084] In one embodiment, the training module is further used to:

[0085] Acquire advertising operation data of different video accounts for playing advertisements from a database, wherein the advertising operation data has a consumption tag;

[0086] The preset recommendation model is trained according to the advertisement operation data with consumption tags to obtain the advertisement recommendation model.

[0087] In one embodiment, the image fusion module 340 is further used for:

[0088] The preset advertisement content replaces the current content of the pixel area to be implanted by using an image synthesis algorithm, and is merged with the key frame image where the pixel area to be implanted is located to obtain a key frame fusion image.

[0089] In one embodiment, the image fusion module 340 is further used for:

[0090] The preset advertisement content is used to cover the current content of the pixel area to be implanted by using an image synthesis algorithm, and is merged with the key frame image where the pixel area to be implanted is located to obtain a key frame fused image.

[0091] like Figure 3 As shown, an embodiment of the present application provides a computer device, including a processor 711, a communication interface 712, a memory 713 and a communication bus 714, wherein the processor 711, the communication interface 712, and the memory 713 communicate with each other through the communication bus 714;

[0092] Memory 713, used for storing computer programs;

[0093] The processor 711 is used to implement the video processing method provided by any one of the aforementioned method embodiments when executing the program stored in the memory 713.

[0094] Those skilled in the art will understand that Figure 3 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.

[0095] In one embodiment, the video processing device provided by the present application can be implemented in the form of a computer program. The computer program can be Figure 3 The computer device shown in the figure is run on the computer device. The memory of the computer device can store various program modules constituting the video processing device, such as: Figure 2 The acquisition module 310, the region identification module 320, the advertisement recommendation module 330, the image fusion module 340 and the output module 350 are shown. The computer program composed of various program modules enables the processor to execute the video processing method of each embodiment of the present application described in this specification.

[0096] Figure 3 The computer device shown can be Figure 2 The acquisition module 310 in the video processing device shown is executed to acquire the currently playing video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account for the historical advertising content. The computer device can perform identification processing on the key frame image to be played in the currently playing video through the area identification module 320 to obtain multiple pixel areas to be implanted. The computer device can perform recommendation of multiple preset advertising contents to the target account based on the historical interaction data through the advertising recommendation module 330. The computer device can perform implantation of the preset advertising content into the pixel area to be implanted through the image fusion module 340, and fuse it with the key frame image where the pixel area to be implanted is located to obtain a key frame fusion image. The computer device can perform, through the output module 350, providing the corresponding key frame fusion image to the target account when the playback frame of the currently playing video reaches the key frame.

[0097] An embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the video processing method provided by any of the aforementioned method embodiments is implemented.

[0098] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0099] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the video processing method described in each embodiment or some parts of the embodiment.

[0100] It should be understood that the terms used herein are only for the purpose of describing specific example embodiments and are not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms "one", "an" and "said" as used herein may also be meant to include plural forms. The terms "include", "comprise", "contain", and "have" are inclusive, and therefore specify the existence of stated features, steps, operations, elements and / or parts, but do not exclude the existence or addition of one or more other features, steps, operations, elements, parts, and / or combinations thereof. The method steps, processes, and operations described herein are not interpreted as necessarily requiring them to be performed in the specific order described or illustrated, unless the execution order is clearly indicated. It should also be understood that additional or alternative can be used.

[0101] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A video processing method, characterized in that: The video processing method comprises: Obtaining the currently played video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account on the historical advertising content; Identify and process the key frame image to be played in the currently played video to obtain a plurality of pixel regions to be implanted; recommending a plurality of preset advertising contents to the target account according to the historical interaction data; Implanting the preset advertising content into the pixel area to be implanted, and fusing it with the key frame image where the pixel area to be implanted is located, to obtain a key frame fused image; When the playback frame of the currently playing video reaches a key frame, the corresponding key frame fusion image is provided to the target account.

2. The video processing method according to claim 1, characterized in that: The step of identifying and processing the key frame image to be played in the currently played video to obtain a plurality of pixel regions to be implanted includes: Identify key frames to be played in the currently playing video to obtain multiple key frame images; An instance segmentation model is used to perform instance segmentation processing on each of the key frame images, and at least one pixel region to be implanted is determined in each of the key frame images.

3. The video processing method according to claim 2, characterized in that: Before obtaining the currently played video and historical interaction data corresponding to the target account, the method further includes: Get multiple key frame images with pixel area selection marks; The preset segmentation model is trained using a plurality of key frame images with pixel region selection marks to obtain the instance segmentation model.

4. The video processing method according to claim 1, characterized in that: The recommending a plurality of preset advertising contents to the target account according to the historical interaction data includes: Inputting the historical interaction data into an advertisement recommendation model, and determining a plurality of advertisement contents to be recommended sorted according to recommendation priorities according to output contents of the advertisement recommendation model; A preset number of advertisement contents to be recommended among the plurality of advertisement contents to be recommended sorted according to recommendation priorities are determined as preset advertisement contents, wherein the preset number is equal to the number of the plurality of pixel areas to be implanted.

5. The video processing method according to claim 4, characterized in that: Before obtaining the currently played video and historical interaction data corresponding to the target account, the method further includes: Acquire advertising operation data of different video accounts for playing advertisements from a database, wherein the advertising operation data has a consumption tag; The preset recommendation model is trained according to the advertisement operation data with consumption tags to obtain the advertisement recommendation model.

6. The video processing method according to claim 1, characterized in that: The step of implanting the preset advertisement content into the pixel region to be implanted and fusing it with the key frame image where the pixel region to be implanted is located to obtain a key frame fused image includes: The preset advertisement content replaces the current content of the pixel area to be implanted by using an image synthesis algorithm, and is merged with the key frame image where the pixel area to be implanted is located to obtain a key frame fusion image.

7. The video processing method according to claim 1, characterized in that: The step of implanting the preset advertisement content into the pixel region to be implanted and fusing it with the key frame image where the pixel region to be implanted is located to obtain a key frame fused image includes: The preset advertisement content is used to cover the current content of the pixel area to be implanted by using an image synthesis algorithm, and is merged with the key frame image where the pixel area to be implanted is located to obtain a key frame fused image.

8. A video processing device, characterized in that: The device comprises: An acquisition module, used to acquire the currently played video and historical interaction data corresponding to the target account, wherein the historical interaction data is used to indicate the operation record of the target account on the historical advertising content; A region recognition module is used to recognize and process the key frame image to be played in the currently played video to obtain a plurality of pixel regions to be implanted; An advertisement recommendation module, configured to recommend a plurality of preset advertisement contents to the target account according to the historical interaction data; An image fusion module, used for implanting the preset advertisement content into the pixel region to be implanted, and fusing it with the key frame image where the pixel region to be implanted is located, to obtain a key frame fusion image; The output module is used to provide the target account with the corresponding key frame fusion image when the playback frame of the currently playing video reaches the key frame.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the video processing method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the video processing method according to any one of claims 1 to 7 is implemented.

Images