Video recommendation method, device, apparatus and storage medium

Abstract

Embodiments of the present application provide a video recommendation method and device, equipment and a storage medium, receiving a video set acquisition request sent by a client, the video set acquisition request comprising a user identifier corresponding to a user; determining a user group category corresponding to the user according to the user identifier; determining a video set corresponding to pre-edited label information matched with the user group category as a target video set, the video set being a set of videos obtained by mixing and cutting a plurality of to-be-cut videos; and sending the target video set to the client to enable the client to play the target video set. Embodiments of the present application enable a target video collection recommended to a user to attract a target user group through video features related to the label information of the target video collection, enhance user viewing stickiness, and improve the propagation effect of the target video, thereby achieving more accurate quick-look video recommendation for different users.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, device, and storage medium for recommending videos. Background Technology

[0002] Existing quick-view video editing methods typically employ multiple editing schemes to split and compress the original long video in order to distribute and convert long video content in short video scenarios.

[0003] However, most existing methods only compress and edit the plot based on the video content itself, failing to adaptively adjust the editing direction based on user feedback during the playback of the original long video. Therefore, they lack targeted adaptation to user feedback and struggle to generate highly viral topics. Furthermore, individual quick-view video collections often cover lengthy plot developments, making it difficult to accurately focus on key or trending content from the original long video, thus hindering more precise quick-view video recommendations for different users.

[0004] Therefore, how to achieve more accurate quick-view video recommendations for different users, so as to improve the distribution effect of quick-view videos and thus enhance user viewing stickiness, is an urgent problem to be solved in this field. Summary of the Invention

[0005] The purpose of this invention is to provide a video recommendation method, apparatus, device, and storage medium to solve the problem of how to achieve more accurate quick-viewing video recommendations for different users, thereby improving the distribution effect of quick-viewing videos and enhancing user viewing stickiness. The specific technical solution is as follows: In a first aspect of the invention, a video recommendation method is provided, applied to a server, the server being communicatively connected to a client, the method comprising: Receive a video collection retrieval request sent by the client, wherein the video collection retrieval request includes the user identifier corresponding to the user; Based on the user identifier, determine the user group category corresponding to the user; The set of videos corresponding to the pre-edited tag information that matches the user group category is determined as the target video set, which is a set of videos obtained by mixing and editing multiple videos to be edited; The target video set is sent to the client so that the client can play the target video set.

[0006] Optionally, the editing method of the video collection includes: Obtain the video to be edited and the key plot information corresponding to the video to be edited. The key plot information is the plot information in the video to be edited that has been used as the reason for the development of the subsequent plot a preset number of times. Based on the key plot information, the creative plot points of the video to be edited are determined, and the creative plot points are the highlights in the video to be edited that are related to the key plot information; The editing direction of the video to be edited is generated based on the creative plot points; The video to be edited is edited according to the editing direction to obtain the video set.

[0007] Optionally, obtaining the key plot information corresponding to the video to be edited includes: Identify the video clips in the video to be edited; Obtain the first introductory message; The video segment and the first introductory text are input into a preset video description generation model to obtain the video description information of the video segment. The first introductory text is used to enable the video description generation model to describe the character behavior characteristics and scene characteristics in the video segment. Obtain the target time granularity, which is the time-based division dimension of the plot description information corresponding to the video to be edited; The shot description information is merged according to the target time granularity to obtain the plot description information corresponding to the video to be edited; The plot description information corresponding to the video to be edited is input into a preset large model so that the large model analyzes the plot description information, determines the plot description information corresponding to the key plot in the video to be edited from the plot description information, takes the plot description information corresponding to the key plot in the video to be edited as the key plot information, and outputs the key plot information.

[0008] Optionally, after obtaining the video to be edited and the key plot information corresponding to the video to be edited, the method includes: Collect the network information corresponding to the video to be edited; The network public opinion topic corresponding to the video to be edited in the network information is located using a pre-set large model. The creative plot points of the video to be edited are determined based on the key plot information and the online public opinion topics. The creative plot points are the plot points in the video to be edited that are related to the key plot information and / or the highlights that are related to the online public opinion topics.

[0009] Optionally, determining the creative plot points of the video to be edited based on the key plot information and the online public opinion topics includes: The key plot information and the online public opinion topics are deduplicated, and the deduplicated key plot information and the online public opinion topics are merged to obtain the creative plot points of the video to be edited; or, The key plot information and the online public opinion topics are input into the large model so that the large model merges the key plot information and the online public opinion topics. Based on the merged key plot information and the online public opinion topics, creative plot points of the video to be edited are generated according to a preset format. The preset format includes at least the creative plot information and corresponding creative highlights of the video to be edited.

[0010] Optionally, generating the editing direction of the video to be edited based on the creative plot points includes: Obtain the second introductory text; The creative plot point and the second guiding text are input into a preset large model; the second guiding text is used to enable the large model to determine the editing direction corresponding to the creative plot point, and to determine the target plot point corresponding to the editing direction in the creative plot point; based on the target plot point, the target user information, plot summary and highlight information corresponding to the editing direction are analyzed; the target user information is the user information corresponding to the target audience of the target plot point. Obtain the editing direction output by the preset large model, wherein the editing direction has the corresponding target user information, plot summary and highlight information.

[0011] Optionally, the step of editing the video to be edited according to the editing direction to obtain the video set includes: Determine the target description information in the plot description information that corresponds to the plot summary in the editing direction; The target description information is bound to the plot summary to obtain a video editing scheme corresponding to the editing direction. The video editing scheme includes plot description information in a preset shot language format, which is used for the screen positioning and video editing of the video to be edited. The video to be edited is edited according to the plot description information in the preset shot language format to obtain the target video; Generate user group profile information corresponding to the target video based on the target user information corresponding to the editing direction; Based on the plot summary and highlight information corresponding to the editing direction, generate video hotspot tag information corresponding to the target video; The target videos are integrated to obtain the video set, and the user group profile information and the video hotspot tag information are used as the tag information of the video set.

[0012] In a second aspect of the invention, a video recommendation device is also provided, applied to a server, the server being communicatively connected to a client, the device comprising: The request receiving module is used to receive a video collection acquisition request sent by the client, wherein the video collection acquisition request includes a user identifier corresponding to the user. The category determination module is used to determine the user group category corresponding to the user based on the user identifier; The set matching module is used to determine the set of videos corresponding to the pre-edited tag information that matches the user group category as the target video set, wherein the video set is a set of videos obtained by mixing and cutting multiple videos to be edited; The video recommendation module is used to send the target video set to the client so that the client can play the target video set.

[0013] In another aspect of the present invention, an electronic device is also provided, characterized in that it includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus. Memory, used to store computer programs; The processor, when executing a program stored in memory, implements the recommended method for any of the above-described videos.

[0014] In another aspect of the invention, a computer-readable storage medium is also provided, wherein instructions are stored therein, which, when executed on a computer, cause the computer to perform the recommended method for the video described above.

[0015] In another aspect of the invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the recommended method for any of the above-described videos.

[0016] This invention provides a video recommendation method that receives a video collection retrieval request from a client, the request including a user identifier corresponding to the user; determines the user group category corresponding to the user based on the user identifier; identifies a video collection with pre-edited tag information matching the user group category as a target video collection, where the video collection is a collection of videos obtained by mixing multiple videos to be edited; and sends the target video collection to the client so that the client can play the target video collection. In this invention, the video collection obtained by mixing multiple videos to be edited has corresponding tag information. When recommending a video collection, the target video collection is determined by matching the tag information of the video collection with the user group category corresponding to the user. This allows the target video collection recommended to the user to attract the target user group through the video characteristics related to its tag information, enhancing user viewing stickiness and improving the dissemination effect of the target videos, thus achieving more accurate quick-view video recommendations for different users. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0018] Figure 1 This is a flowchart illustrating the steps of a video recommendation method embodiment of the present invention. Figure 2 This is a schematic diagram of the video editing process of an embodiment of the video recommendation method of the present invention; Figure 3 This is a structural block diagram of an embodiment of a video recommendation device according to the present invention; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the present invention will now be described with reference to the accompanying drawings in the embodiments of the present invention.

[0020] Reference Figure 1 The above is a flowchart of an embodiment of a video recommendation method according to the present invention, including the following steps: Step 101: Receive the video collection acquisition request sent by the client. The video collection acquisition request includes the user identifier corresponding to the user.

[0021] This invention provides a video recommendation method applied to a server, which communicates with a client. Users can perform operations such as searching, playing, and watching videos through the client, which can recommend and play videos to users. For example, users can watch videos recommended by the client through the video recommendation function in an application installed on the client. The video played by the client can be a target video from a video collection.

[0022] In this embodiment of the invention, the video set is a quick-view video set, which contains one or more quick-view videos obtained through cross-editing. The video set is a collection of quick-view videos obtained by cross-editing one or more videos to be edited.

[0023] In one embodiment, video editing can shorten one or more longer videos into one or more shorter target videos. In an optional embodiment of the invention, the video to be edited can be a long video, and the target video can be a quick-view video. The long video can be a long video album containing multiple episodes, such as a TV series with multiple episodes. A quick-view video is a short video used to quickly explain or showcase the content of a corresponding long video in a very short time; the duration of a quick-view video is usually shorter than the duration of the corresponding long video. In some cases, a quick-view video can also contain the content of multiple long videos. For example, if a TV series contains 20 episodes, two quick-view videos can be edited to correspond to episodes 1-10 and 11-20 of the TV series, respectively.

[0024] In step 101, the client needs to obtain the target video set recommended to the user from the server and sends a video set retrieval request to the server, including the user's corresponding user identifier. The server receives the video set retrieval request sent by the client and determines the target video set for recommendation based on the user identifier. For example, when a user uses the video recommendation function in an application installed on the client, it can trigger the client to send a video set retrieval request to the server, including the user's corresponding user identifier. This embodiment of the invention does not limit the method of triggering the client to send a video set retrieval request to the server.

[0025] Step 102: Determine the user group category corresponding to the user based on the user identifier.

[0026] In step 102, the server obtains the user identifier from the request based on the video set, and can determine the user group category corresponding to that user. The user group category is a user group divided according to the user's characteristics. For example, the user group category may include, but is not limited to, user groups divided according to characteristics such as the user's gender, age, and video preference type.

[0027] Step 103: Determine the set of videos corresponding to the pre-edited tag information that matches the user group category as the target video set. The video set is a set of videos obtained by mixing and editing multiple videos to be edited.

[0028] In this embodiment of the invention, the video collection obtained by mixing multiple videos to be edited all have corresponding tag information. In one embodiment, the tag information includes user group profile information and video hotspot tag information corresponding to the videos in the video collection. The user group profile information refers to the characteristics of the main audience user group of the videos in the video collection, such as the user's age and gender. The video hotspot tag information consists of keywords that the video collection may become hotspots on the dissemination platform. Tagging the target video collection with hotspot tags can increase the probability of being recommended by the system and discovered by target users. The video hotspot tag information may include the plot summary and highlights of the video.

[0029] In step 103, the server determines the set of videos corresponding to pre-edited tag information that matches the user group category as the target video set. Specifically, when matching the target video set, the server can match user group profile information and / or video hotspot tag information with the user group category to obtain the target video set.

[0030] Step 104: Send the target video set to the client so that the client can play the target video set.

[0031] In step 104, the server sends the determined target video set to the client. After receiving the target video set, the client recommends or plays the video, thereby recommending the target video set to the user.

[0032] In this embodiment of the invention, a video set with corresponding tag information is obtained by mixing and editing multiple videos to be edited. When recommending the video set, the tag information of the video set is matched with the user's corresponding user group category to determine the target video set for recommendation. This allows the target video set recommended to the user to attract the target user group through the video characteristics related to its tag information, enhance user viewing stickiness, and improve the dissemination effect of the target video, thereby achieving more accurate quick-watch video recommendations for different users.

[0033] Optionally, the editing method of the video collection includes: Obtain the video to be edited and the key plot information corresponding to the video to be edited. The key plot information is the plot information in the video to be edited that has been used as the reason for the development of the subsequent plot a preset number of times. Based on the key plot information, the creative plot points of the video to be edited are determined, and the creative plot points are the highlights in the video to be edited that are related to the key plot information; The editing direction of the video to be edited is generated based on the creative plot points; The video to be edited is edited according to the editing direction to obtain the video set.

[0034] This invention, in its embodiments, edits a video to be edited based on key plot information, resulting in a video collection. Specifically, it acquires the video to be edited into the target video and extracts its key plot information. Key plot information refers to the set of plot elements in the video to be edited that play a core role in plot progression, character development, and the expression of the video's theme; it is plot information that has served as a cause for subsequent plot development a preset number of times in the video to be edited.

[0035] In one embodiment, if the video to be edited is a TV series with multiple episodes, the TV series can be processed using video description generation technology, such as a multimodal video understanding model. The multimodal video understanding model is used to understand the video content of the TV series and generate plot description information for the TV series, thereby identifying the key plot information of the video to be edited from the plot description information.

[0036] Key plot information is organized to obtain several creative plot points for the video to be edited. In one embodiment, a large model can be used to organize the key plot information and generate creative plot points for the video to be edited according to a preset format. Creative plot points are highlights in the video to be edited that are related to the key plot information. They can include the key plot information and the corresponding creative highlights, serving as a basis for the editing direction of the video. For example, if the key plot information is how the protagonist and his lover meet, then the corresponding creative highlight could be the emotional entanglement between the protagonist and his lover. The creative plot points organize different key plot information and the creative highlights corresponding to each key plot information.

[0037] By processing different creative plot points in the video to be edited, several editing directions can be obtained. Each editing direction can revolve around one or more creative plot points, and the creative plot point revolving around each editing direction is the target plot point of that editing direction. For example, if the video to be edited is a TV series, one editing direction can revolve around the growth experience of the main character in the TV series, another editing direction can revolve around the emotional entanglement between the main character and his lover, and so on, thus obtaining multiple editing directions corresponding to the TV series. In one embodiment, a large model can be used to generate the editing directions of the video to be edited based on the creative plot points.

[0038] In one embodiment, an editing direction may include a corresponding target plot point and editing information, the editing information being used to determine the target video's audience, plot summary, and highlights, etc., obtained by editing according to the editing direction.

[0039] For each editing direction, a corresponding video editing scheme is generated. Each editing direction can generate one or more video editing schemes, and each video editing scheme corresponds to a target video for that editing direction. Based on the video editing scheme for each editing direction, the video to be edited is edited separately, resulting in a video collection for each editing direction. The video collections obtained for each different editing direction can be used as dimensions for subsequent short video recommendation and distribution. Each video collection includes the target videos edited according to each video editing scheme for that editing direction.

[0040] In one embodiment, existing quick-view video generation technology can be used to process each video editing scheme to obtain the corresponding target video, thereby generating a corresponding video set for each editing direction.

[0041] This invention enables the determination of creative plot points in a video to be edited based on key plot information. Since the key plot information contains crucial storylines, the determined creative plot points accurately focus on the key content of the original video, allowing the final video collection to attract target audiences by focusing on these key plot points, enhancing user engagement and improving the video's dissemination. Furthermore, this invention can generate different editing directions based on the creative plot points, and then edit the video based on each direction to obtain video collections with varying styles. This provides support for the reserve of target video content for different user groups, effectively enhancing the content diversity of the target videos.

[0042] Optionally, obtaining the key plot information corresponding to the video to be edited includes: Identify the video clips in the video to be edited; Obtain the first introductory message; The video segment and the first introductory text are input into a preset video description generation model to obtain the video description information of the video segment. The first introductory text is used to enable the video description generation model to describe the character behavior characteristics and scene characteristics in the video segment. Obtain the target time granularity, which is the time-based division dimension of the plot description information corresponding to the video to be edited; The shot description information is merged according to the target time granularity to obtain the plot description information corresponding to the video to be edited; The plot description information corresponding to the video to be edited is input into a preset large model so that the large model analyzes the plot description information, determines the plot description information corresponding to the key plot in the video to be edited from the plot description information, takes the plot description information corresponding to the key plot in the video to be edited as the key plot information, and outputs the key plot information.

[0043] In this embodiment of the invention, a preset video description generation model can be used to understand the content of the video to be edited, generate plot description information corresponding to different time granularities, and then further analyze the plot description information to obtain the key plot information of the video to be edited.

[0044] Specifically, the video segments in the video to be edited are identified. All video frames in a segment correspond to the same shot, which is a continuous sequence of footage captured by the camera from startup to shutdown. The video segments and a first prompt are input into a preset video description generation model. The first prompt serves as a cue for the model and may include instructions to enable it to comprehensively understand the video segments and describe the behavioral and scene characteristics within them. Thus, the video description generation model can comprehensively understand the video segments based on the first prompt and describe the behavioral and scene characteristics to obtain the shot description information corresponding to each video segment.

[0045] Obtaining plot description information requires a corresponding target time granularity. The target time granularity is the time-based division dimension of the plot description information corresponding to the video to be edited. For example, the target time granularity can be a segment, scene, series, album, etc., and the corresponding plot description information is respectively the plot description information of each segment of the video to be edited, the plot description information of each scene of the video to be edited, the plot description information corresponding to each series of the video to be edited, and the plot description information corresponding to the entire video to be edited, etc.

[0046] The shot description information is merged according to the target time granularity to obtain the plot description information corresponding to different target time granularities. For example, if the target time granularity is scene, the shot description information of the shot video segments corresponding to each scene in the video to be edited will be merged to obtain the plot description information at the scene level.

[0047] Analyzing the plot description information can yield key plot information corresponding to the video to be edited. In one embodiment, plot description information at the album level can be analyzed, that is, the plot description information corresponding to the entire video to be edited, to obtain the key plot information corresponding to the video to be edited.

[0048] Specifically, the plot description information at the album level can be input into a preset large model, and corresponding prompts can be set. The large model can then comprehensively understand, summarize, and analyze the plot description information at the album level based on the prompts. From the plot description information, the model can identify the plot description information that plays a core role in the plot of the video to be edited, such as plot progression, character development, and the expression of the video theme. From these plot description information, the model can identify the plot information that has been used as the cause of subsequent plot development in the video to be edited a preset number of times, and output it as the key plot information corresponding to the video to be edited.

[0049] This invention identifies the video content of the video to be edited and generates corresponding plot description information, thereby obtaining key plot information of the video to be edited. This enables the understanding of the video to be edited from the perspective of video content analysis. By extracting key plot information from the plot description information of the video to be edited through a large model, the key plot information is used to determine creative plot points and editing direction, so that the target video obtained by subsequent editing can accurately focus on the key content of the original video to be edited.

[0050] Optionally, after obtaining the video to be edited and the key plot information corresponding to the video to be edited, the method includes: Collect the network information corresponding to the video to be edited; The network public opinion topic corresponding to the video to be edited in the network information is located using a pre-set large model. The creative plot points of the video to be edited are determined based on the key plot information and the online public opinion topics. The creative plot points are the plot points in the video to be edited that are related to the key plot information and / or the highlights that are related to the online public opinion topics.

[0051] In this embodiment of the invention, in order for the final generated collection of target videos to attract the target user group by focusing on key plot points and public opinion topics, it is necessary to understand the videos to be edited from two dimensions: the plot content description of the videos to be edited and the capture of online trending information. Therefore, in addition to the key plot information of the videos to be edited obtained in the above embodiments, it is also necessary to obtain the corresponding online public opinion topics of the videos to be edited.

[0052] Specifically, a large-scale model is used to collect online information related to the video to be edited, or a search engine is used to crawl online information related to the video to be edited. The large-scale model can be one that supports text modality understanding, and the channels for collecting online information can include various channels such as search engines and social media. The online information can be trending topics related to the video to be edited. After collecting the online information related to the video to be edited, the large-scale model is used to summarize and mine the online information, locating the public opinion topics related to the plot of the video to be edited, and these public opinion topics are taken as the online public opinion topics corresponding to the video to be edited.

[0053] In one embodiment, a corresponding prompt word is set to input a large model. The prompt word may include a command to use the large model to collect network information corresponding to the video to be edited on the network, and to summarize and mine the information to be edited to locate the public opinion topics related to the plot of the information to be edited contained therein. This allows the large model to perform relevant operations according to the command of the prompt word to obtain the online public opinion topics corresponding to the video to be edited.

[0054] The process involves integrating and processing key plot information and online public opinion topics to summarize these elements, thereby generating several creative plot points for the video to be edited. These creative plot points are highlights in the video that relate to the key plot information and / or to the online public opinion topics. These creative plot points are the main plot highlights that attract the target audience and can be used as a basis for the editing direction of the video.

[0055] This invention can determine the creative plot points of a video to be edited by combining key plot information and online public opinion topics. Since the key plot information contains the key plot points of the video to be edited, and the online public opinion topics contain user feedback on the video to be edited, the determined creative plot points can accurately focus on the key content of the original video to be edited, and also cover the hot topics within the video to be edited. This allows the final generated collection of target videos to attract the target user group by focusing on key plot points and public opinion topics, enhancing user engagement and improving the dissemination effect of the target videos. Optionally, determining the creative plot points of the video to be edited based on the key plot information and the online public opinion topics includes: The key plot information and the online public opinion topics are deduplicated, and the deduplicated key plot information and the online public opinion topics are merged to obtain the creative plot points of the video to be edited; or, The key plot information and the online public opinion topics are input into the large model so that the large model merges the key plot information and the online public opinion topics. Based on the merged key plot information and the online public opinion topics, creative plot points of the video to be edited are generated according to a preset format. The preset format includes at least the creative plot information and corresponding creative highlights of the video to be edited.

[0056] In this embodiment of the invention, after obtaining the key plot information and online public opinion topics of the video to be edited, it is necessary to summarize the key plot information and online public opinion topics to generate creative plot points.

[0057] Specifically, embodiments of the present invention can obtain the creative plot points of the video to be edited by deduplicating key plot information and online opinion topics, and then merging the deduplicated key plot information and online opinion topics. This is a relatively simple summarization method.

[0058] Alternatively, this embodiment of the invention also provides another aggregation method. By setting corresponding prompts, key plot information and online public opinion topics are input into a large model, so that the preset large model merges the key plot information and online public opinion topics according to the prompts, and generates creative plot points of the video to be edited according to a preset format based on the merged key plot information and online public opinion topics. The prompts include commands that instruct the large model to merge key plot information and online public opinion topics, and generate creative plot points of the video to be edited according to a preset format based on the merged key plot information and online public opinion topics.

[0059] In one embodiment, the creative plot points of the video to be edited may include corresponding plot information and creative highlights. Creative highlights refer to the features of the creative plot points that are most likely to attract target users to watch, trigger interaction, or spread the word; they are generally the most appealing plot features within the creative plot points. Therefore, the preset format includes at least the creative plot information and corresponding creative highlights of the video to be edited.

[0060] This invention summarizes key plot information and online public opinion topics from the video to be edited to obtain creative plot points. This allows for a comprehensive consideration of key and trending content in the original video to be edited when generating editing directions based on these creative plot points.

[0061] Optionally, generating the editing direction of the video to be edited based on the creative plot points includes: Obtain the second introductory text; The creative plot point and the second guiding text are input into a preset large model; the second guiding text is used to enable the large model to determine the editing direction corresponding to the creative plot point, and to determine the target plot point corresponding to the editing direction in the creative plot point; based on the target plot point, the target user information, plot summary and highlight information corresponding to the editing direction are analyzed; the target user information is the user information corresponding to the target audience of the target plot point. Obtain the editing direction output by the preset large model, wherein the editing direction has the corresponding target user information, plot summary and highlight information.

[0062] In this embodiment of the invention, an editing direction for the video to be edited is generated around the creative plot points of the video to be edited. An editing direction refers to extracting a specific theme for editing the target video content based on the key plot points and public opinion topics of the video to be edited, serving as the core framework for subsequent editing of the target video. An editing direction can revolve around one or more creative plot points.

[0063] Specifically, the creative plot points and the second guiding phrase are input into the large model, enabling the preset large model to determine the target plot point corresponding to each editing direction from the creative plot points. This editing direction will then revolve around the target plot point. The second guiding phrase is a prompt word input into the large model, serving as a command to the large model to determine the editing directions that can be selected for the creative plot point, and to identify the target plot points corresponding to different editing directions within the creative plot point. Each clip can currently correspond to one or more target plot points. Based on the target plot points, the target user information, plot summary, and highlight information corresponding to each editing direction are analyzed.

[0064] The target user information refers to the user information corresponding to the target audience of the target plot point, that is, the main audience group of the target video for this editing direction. The plot summary is the plot summary of the target plot point for this editing direction. Highlight information can include creative titles and creative highlights. Creative titles are used to identify different editing directions, and creative highlights are the creative highlights corresponding to the target plot point of this editing direction. In other words, the instructions of the second guide need to make the output of the large model clearly state the target audience of this editing direction, the corresponding title of this editing direction, the editing characteristics of this editing direction, and the brief plot information included in this editing direction.

[0065] Therefore, each edit direction output by the final large model includes the corresponding target plot point, target user information, plot summary, and highlight information. Organizing the multiple edit directions output by the large model yields a set of edit directions.

[0066] This invention uses a large model to process creative plot points, resulting in different editing directions including target plot points, target user information, plot summary, and highlight information. These directions are then used to edit the videos to be edited, resulting in video collections with different styles. This provides support for the reserve of target video content for different user groups and effectively improves the content diversity of target videos.

[0067] Optionally, the step of editing the video to be edited according to the editing direction to obtain the video set includes: Determine the target description information in the plot description information that corresponds to the plot summary in the editing direction; The target description information is bound to the plot summary to obtain a video editing scheme corresponding to the editing direction. The video editing scheme includes plot description information in a preset shot language format, which is used for the screen positioning and video editing of the video to be edited. The video to be edited is edited according to the plot description information in the preset shot language format to obtain the target video; Generate user group profile information corresponding to the target video based on the target user information corresponding to the editing direction; Based on the plot summary and highlight information corresponding to the editing direction, generate video hotspot tag information corresponding to the target video; The target videos are integrated to obtain the video set, and the user group profile information and the video hotspot tag information are used as the tag information of the video set.

[0068] In this embodiment of the invention, after filling in the details for each editing direction, one or more video editing schemes corresponding to each editing direction can be obtained. Each video editing scheme corresponds to a target video. The video to be edited is edited according to the video editing scheme to obtain the target video. The target videos are then integrated to obtain a video set.

[0069] Specifically, combining the plot description information corresponding to different target time granularities generated in the above embodiments, the large model can locate target description information related to the plot summary corresponding to the editing direction within the plot description information corresponding to different target time granularities. After binding these target description information with the plot summary corresponding to the editing direction, detail filling can be achieved. After performing detail filling, the large model further converts the bound target description information and the corresponding plot summary into plot description information in a shot language format containing temporal information. The plot description information in a shot language format containing temporal information can be directly used for screen positioning and video editing of the video to be edited. Among them, the plot description information in a shot language format containing temporal information can include detailed shot language format description information describing the character behavior and scene in the video to be edited.

[0070] The large model organizes the plot description information corresponding to each editing direction to obtain the video editing scheme for that editing direction. Each editing direction corresponds to a set of videos. Depending on the plot development of the specific video to be edited, an editing direction can contain one or more video editing schemes, and each video editing scheme corresponds to a target video.

[0071] In one embodiment, by setting corresponding prompt words, the prompt words include a command that enables the large model to determine the target description information in the plot description information that corresponds to the plot summary of the editing direction, and bind the target description information with the plot summary to obtain the video editing scheme corresponding to the editing direction. The prompt words are input into the large model so that the large model can perform the operation of determining the target description information and generating the video editing scheme corresponding to the editing direction.

[0072] Using existing quick-view video generation technology, the video to be edited is edited according to the plot description information in the preset shot language format of each video editing scheme to obtain the target video. One or more target videos corresponding to each editing direction are then integrated into a target video set corresponding to that editing direction.

[0073] In this embodiment of the invention, each video set should also include corresponding tag information for use in the subsequent video recommendation matching process. The tag information includes user profile information of the video set and video trending tag information, etc.

[0074] The user profile information refers to the main target audience of the videos in the video collection, derived from the target user information of the editing direction corresponding to the video collection. The video trend tag information consists of keywords that the video collection is likely to become trending on the distribution platform. Tagging the target video collection with trend tags can increase the probability of being recommended by the system and discovered by target users. The video trend tag information is derived from the plot summary and highlight information of the editing direction corresponding to the video collection.

[0075] This invention, through a pre-set large model, uses plot description information corresponding to the target time granularity to fill in the details of the plot summary for the editing direction, thereby automatically generating a video editing scheme for the editing of the target video. This invention also sets corresponding user group profile information and video hotspot tag information for each set of target videos, facilitating attracting user interest and improving the dissemination effect of the target videos during subsequent platform distribution.

[0076] To enable those skilled in the art to more clearly understand the video collection editing process in a video recommendation method provided in this embodiment of the invention, the following describes... Figure 2 The editing process of a video collection in a video recommendation method provided by an embodiment of the present invention will be explained.

[0077] Reference Figure 2 This is a schematic diagram of the video editing process of an embodiment of the video recommendation method of the present invention.

[0078] like Figure 2 As shown, the video to be edited can be a TV series containing multiple episodes, with each episode being a long video.

[0079] To ensure the final video compilation attracts the target audience by focusing on key plot points and trending topics within the videos to be edited, the process involves understanding the videos from two dimensions: plot description and online trending topics. Specifically, for the plot description dimension, a pre-defined video description generation model is used to understand the content of the videos to be edited, generating plot description information at different time granularities. The plot description information at the album level is used as the long-form video plot description for the corresponding TV series, and key plot information is obtained by analyzing this long-form video plot description. For the online trending topics dimension, a large-scale model is used to collect online trending topics related to the videos to be edited. This large-scale model is then used to summarize and mine the online information, identifying relevant trending topics related to the plot of the videos to be edited. These trending topics are then used as the corresponding online trending topics for the videos to be edited.

[0080] Key plot information and online discussion topics are summarized to obtain several creative plot points for the video to be edited. Further, a set of editing directions is generated based on these creative plot points, serving as a quick-view creative set for the video. After filling the quick-view creative set with plot details using plot description information corresponding to various time granularities, a video editing scheme corresponding to each editing direction for the video to be edited is obtained.

[0081] The video to be edited is edited according to the video editing scheme corresponding to each editing direction, resulting in quick-view videos (i.e., target videos) for each editing direction. User profile information, video trending tags, and quick-view video albums corresponding to that editing direction are then integrated to obtain a quick-view video collection for that editing direction. Ultimately, a TV series can be edited into multiple quick-view video collections corresponding to different editing directions, and these collections will be subsequently recommended to users.

[0082] This invention can combine key plot information of the video to be edited with online public opinion topics to determine the creative plot points. Since the key plot information contains the key plot points of the video to be edited, and the online public opinion topics contain user feedback, the determined creative plot points can accurately focus on the key content of the original video to be edited, while also covering the hot topics within it. This allows the final generated collection of target videos to attract the target audience by focusing on key plot points and public opinion topics, enhancing user engagement and improving the dissemination effect of the target videos. Furthermore, this invention can generate different editing directions based on the creative plot points, and then edit the video to be edited based on each editing direction to obtain target video collections with different styles. This can support the reserve of target video content for different user groups, effectively improving the content diversity of the target videos.

[0083] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0084] Based on the above embodiments, this embodiment also provides a video recommendation device, which can be applied to electronic devices such as terminal devices and servers.

[0085] Reference Figure 3 The diagram illustrates a structural block diagram of an embodiment of a video recommendation device according to the present invention, which may specifically include the following modules: The request receiving module 301 is used to receive a video collection acquisition request sent by the client, wherein the video collection acquisition request includes a user identifier corresponding to the user. The category determination module 302 is used to determine the user group category corresponding to the user based on the user identifier; The set matching module 303 is used to determine the set of videos corresponding to the pre-edited tag information that matches the user group category as the target video set, wherein the video set is a set of videos obtained by mixing and cutting multiple videos to be edited; The video recommendation module 304 is used to send the target video set to the client so that the client can play the target video set.

[0086] Optionally, the device further includes: The key plot information acquisition module is used to acquire the video to be edited and the key plot information corresponding to the video to be edited. The key plot information is plot information in the video to be edited that has been used as the reason for the development of the subsequent plot a preset number of times. The first creative plot point determination module is used to determine the creative plot points of the video to be edited based on the key plot information. The creative plot points are highlights in the video to be edited that are related to the key plot information. The editing direction generation module is used to generate the editing direction of the video to be edited based on the creative plot points; The video editing module is used to edit the video to be edited according to the editing direction to obtain the video set.

[0087] Optionally, the key plot information acquisition module includes: The shot video segment recognition submodule is used to identify shot video segments in the video to be edited; The first introductory phrase retrieval submodule is used to retrieve the first introductory phrase. The shot description information determination submodule is used to input the shot video segment and the first prompt into a preset video description generation model to obtain the shot description information of the shot video segment. The first prompt is used to enable the video description generation model to describe the character behavior characteristics and scene characteristics in the shot video segment. The time granularity acquisition submodule is used to acquire the target time granularity, which is the time-based division dimension of the plot description information corresponding to the video to be edited; The shot description information merging submodule is used to merge the shot description information according to the target time granularity to obtain the plot description information corresponding to the video to be edited. The key plot information determination submodule is used to input the plot description information corresponding to the video to be edited into a preset large model, so that the large model analyzes the plot description information, determines the plot description information corresponding to the key plot in the video to be edited from the plot description information, takes the plot description information corresponding to the key plot in the video to be edited as the key plot information, and outputs the key plot information.

[0088] Optionally, the device further includes: A network information collection module is used to collect network information corresponding to the video to be edited; The online public opinion topic identification module is used to locate the online public opinion topic corresponding to the video to be edited in the online information using a preset large model; The second creative plot point determination module is used to determine the creative plot points of the video to be edited based on the key plot information and the online public opinion topic. The creative plot points are plot points in the video to be edited that are related to the key plot information and / or highlight information that is related to the online public opinion topic.

[0089] Optionally, the second creative sentiment point determination module includes: The first creative plot point generation submodule is used to deduplicatize the key plot information and the online public opinion topic points, and merge the deduplicated key plot information and the online public opinion topic points to obtain the creative plot points of the video to be edited. or, The second creative plot point generation submodule is used to input the key plot information and the online public opinion topic into the large model, so that the large model merges the key plot information and the online public opinion topic, and generates creative plot points of the video to be edited according to the merged key plot information and the online public opinion topic in a preset format. The preset format includes at least the creative plot information and corresponding creative highlights of the video to be edited.

[0090] Optionally, the clipping direction generation module includes: The second introductory text acquisition submodule is used to acquire the second introductory text. The first editing direction determination submodule is used to input the creative plot point and the second guiding text into a preset large model; the second guiding text is used to enable the large model to determine the editing direction corresponding to the creative plot point, and to determine the target plot point corresponding to the editing direction in the creative plot point; based on the target plot point, the target user information, plot summary and highlight information corresponding to the editing direction are analyzed; the target user information is the user information corresponding to the target audience of the target plot point. The second editing direction determination submodule is used to obtain the editing direction output by the preset large model. The editing direction has the corresponding target user information, plot summary and highlight information.

[0091] Optionally, the video editing module includes: The target description information determination submodule is used to determine the target description information in the plot description information that corresponds to the plot summary of the editing direction; The video editing scheme determination submodule is used to bind the target description information with the plot summary to obtain the video editing scheme corresponding to the editing direction. The video editing scheme includes plot description information in a preset shot language format, which is used for the screen positioning and video editing of the video to be edited. The target video editing submodule is used to edit the video to be edited according to the plot description information in the preset shot language format to obtain the target video; The user group profile information determination submodule is used to generate user group profile information corresponding to the target video based on the target user information corresponding to the editing direction. The video hotspot tag information determination submodule is used to generate video hotspot tag information corresponding to the target video based on the plot summary and the highlight information corresponding to the editing direction. The video collection integration submodule is used to integrate the target videos to obtain the video collection, and to use the user group profile information and the video hotspot tag information as the tag information of the video collection.

[0092] This invention also provides an electronic device, such as... Figure 4 As shown, it includes a processor 401, a communication interface 402, a memory 403, and a communication bus 404, wherein the processor 401, the communication interface 402, and the memory 403 communicate with each other through the communication bus 404. Memory 403 is used to store computer programs; When processor 401 executes the program stored in memory 403, it performs the following steps: Receive a video collection retrieval request sent by the client, wherein the video collection retrieval request includes the user identifier corresponding to the user; Based on the user identifier, determine the user group category corresponding to the user; The set of videos corresponding to the pre-edited tag information that matches the user group category is determined as the target video set, which is a set of videos obtained by mixing and editing multiple videos to be edited; The target video set is sent to the client so that the client can play the target video set.

[0093] Optionally, the editing method of the video collection includes: Obtain the video to be edited and the key plot information corresponding to the video to be edited. The key plot information is the plot information in the video to be edited that has been used as the reason for the development of the subsequent plot a preset number of times. Based on the key plot information, the creative plot points of the video to be edited are determined, and the creative plot points are the highlights in the video to be edited that are related to the key plot information; The editing direction of the video to be edited is generated based on the creative plot points; The video to be edited is edited according to the editing direction to obtain the video set.

[0094] Optionally, obtaining the key plot information corresponding to the video to be edited includes: Identify the video clips in the video to be edited; Obtain the first introductory message; The video segment and the first introductory text are input into a preset video description generation model to obtain the video description information of the video segment. The first introductory text is used to enable the video description generation model to describe the character behavior characteristics and scene characteristics in the video segment. Obtain the target time granularity, which is the time-based division dimension of the plot description information corresponding to the video to be edited; The shot description information is merged according to the target time granularity to obtain the plot description information corresponding to the video to be edited; The plot description information corresponding to the video to be edited is input into a preset large model so that the large model analyzes the plot description information, determines the plot description information corresponding to the key plot in the video to be edited from the plot description information, takes the plot description information corresponding to the key plot in the video to be edited as the key plot information, and outputs the key plot information.

[0095] Optionally, after obtaining the video to be edited and the key plot information corresponding to the video to be edited, the method includes: Collect the network information corresponding to the video to be edited; The network public opinion topic corresponding to the video to be edited in the network information is located using a pre-set large model. The creative plot points of the video to be edited are determined based on the key plot information and the online public opinion topics. The creative plot points are the plot points in the video to be edited that are related to the key plot information and / or the highlights that are related to the online public opinion topics.

[0096] Optionally, determining the creative plot points of the video to be edited based on the key plot information and the online public opinion topics includes: The key plot information and the online public opinion topics are deduplicated, and the deduplicated key plot information and the online public opinion topics are merged to obtain the creative plot points of the video to be edited; or, The key plot information and the online public opinion topics are input into the large model so that the large model merges the key plot information and the online public opinion topics. Based on the merged key plot information and the online public opinion topics, creative plot points of the video to be edited are generated according to a preset format. The preset format includes at least the creative plot information and corresponding creative highlights of the video to be edited.

[0097] Optionally, generating the editing direction of the video to be edited based on the creative plot points includes: Obtain the second introductory text; The creative plot point and the second guiding text are input into a preset large model; the second guiding text is used to enable the large model to determine the editing direction corresponding to the creative plot point, and to determine the target plot point corresponding to the editing direction in the creative plot point; based on the target plot point, the target user information, plot summary and highlight information corresponding to the editing direction are analyzed; the target user information is the user information corresponding to the target audience of the target plot point. Obtain the editing direction output by the preset large model, wherein the editing direction has the corresponding target user information, plot summary and highlight information.

[0098] Optionally, the step of editing the video to be edited according to the editing direction to obtain the video set includes: Determine the target description information in the plot description information that corresponds to the plot summary in the editing direction; The target description information is bound to the plot summary to obtain a video editing scheme corresponding to the editing direction. The video editing scheme includes plot description information in a preset shot language format, which is used for the screen positioning and video editing of the video to be edited. The video to be edited is edited according to the plot description information in the preset shot language format to obtain the target video; Generate user group profile information corresponding to the target video based on the target user information corresponding to the editing direction; Based on the plot summary and highlight information corresponding to the editing direction, generate video hotspot tag information corresponding to the target video; The target videos are integrated to obtain the video set, and the user group profile information and the video hotspot tag information are used as the tag information of the video set.

[0099] The communication bus mentioned above can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0100] The communication interface is used for communication between the aforementioned terminal and other devices.

[0101] The memory may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0102] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0103] In another embodiment of the present invention, a computer-readable storage medium is also provided, which stores instructions that, when executed on a computer, cause the computer to perform the recommended method for video as described in any of the above embodiments.

[0104] In yet another embodiment of the present invention, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute the video recommendation method described in any of the above embodiments.

[0105] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk (SSD)).

[0106] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0107] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0108] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A video recommendation method, characterized in that, Applied to a server, wherein the server communicates with a client, the method includes: Receive a video collection retrieval request sent by the client, wherein the video collection retrieval request includes the user identifier corresponding to the user; Based on the user identifier, determine the user group category corresponding to the user; The set of videos corresponding to the pre-edited tag information that matches the user group category is determined as the target video set, which is a set of videos obtained by mixing and editing multiple videos to be edited; The target video set is sent to the client so that the client can play the target video set.

2. The video recommendation method according to claim 1, characterized in that, The editing methods for the video collection include: Obtain the video to be edited and the key plot information corresponding to the video to be edited. The key plot information is the plot information in the video to be edited that has been used as the reason for the development of the subsequent plot a preset number of times. Based on the key plot information, the creative plot points of the video to be edited are determined, and the creative plot points are the highlights in the video to be edited that are related to the key plot information; The editing direction of the video to be edited is generated based on the creative plot points; The video to be edited is edited according to the editing direction to obtain the video set.

3. The video recommendation method according to claim 2, characterized in that, The step of obtaining the key plot information corresponding to the video to be edited includes: Identify the video clips in the video to be edited; Obtain the first introductory message; The video segment and the first introductory text are input into a preset video description generation model to obtain the video description information of the video segment. The first introductory text is used to enable the video description generation model to describe the character behavior characteristics and scene characteristics in the video segment. Obtain the target time granularity, which is the time-based division dimension of the plot description information corresponding to the video to be edited; The shot description information is merged according to the target time granularity to obtain the plot description information corresponding to the video to be edited; The plot description information corresponding to the video to be edited is input into a preset large model so that the large model analyzes the plot description information, determines the plot description information corresponding to the key plot in the video to be edited from the plot description information, takes the plot description information corresponding to the key plot in the video to be edited as the key plot information, and outputs the key plot information.

4. The video recommendation method according to claim 2, characterized in that, After obtaining the video to be edited and the corresponding key plot information, the method includes: Collect the network information corresponding to the video to be edited; The network public opinion topic corresponding to the video to be edited in the network information is located using a pre-set large model. The creative plot points of the video to be edited are determined based on the key plot information and the online public opinion topics. The creative plot points are the plot points in the video to be edited that are related to the key plot information and / or the highlights that are related to the online public opinion topics.

5. The video recommendation method according to claim 4, characterized in that, The process of determining the creative plot points of the video to be edited based on the key plot information and the online public opinion topics includes: The key plot information and the online public opinion topics are deduplicated, and the deduplicated key plot information and the online public opinion topics are merged to obtain the creative plot points of the video to be edited; or, The key plot information and the online public opinion topics are input into the large model so that the large model merges the key plot information and the online public opinion topics. Based on the merged key plot information and the online public opinion topics, creative plot points of the video to be edited are generated according to a preset format. The preset format includes at least the creative plot information and corresponding creative highlights of the video to be edited.

6. The video recommendation method according to any one of claims 2-5, characterized in that, The step of generating the editing direction of the video to be edited based on the creative sentiment points includes: Obtain the second introductory text; The creative plot point and the second guiding text are input into a preset large model; the second guiding text is used to enable the large model to determine the editing direction corresponding to the creative plot point, and to determine the target plot point corresponding to the editing direction in the creative plot point; based on the target plot point, the target user information, plot summary and highlight information corresponding to the editing direction are analyzed; the target user information is the user information corresponding to the target audience of the target plot point. Obtain the editing direction output by the preset large model, wherein the editing direction has the corresponding target user information, plot summary and highlight information.

7. The video recommendation method according to claim 6, characterized in that, The process of editing the video to be edited according to the editing direction to obtain the video set includes: Determine the target description information in the plot description information that corresponds to the plot summary in the editing direction; The target description information is bound to the plot summary to obtain a video editing scheme corresponding to the editing direction. The video editing scheme includes plot description information in a preset shot language format, which is used for the screen positioning and video editing of the video to be edited. The video to be edited is edited according to the plot description information in the preset shot language format to obtain the target video; Generate user group profile information corresponding to the target video based on the target user information corresponding to the editing direction; Based on the plot summary and highlight information corresponding to the editing direction, generate video hotspot tag information corresponding to the target video; The target videos are integrated to obtain the video set, and the user group profile information and the video hotspot tag information are used as the tag information of the video set.

8. A video recommendation device, characterized in that, Applied to a server, the server communicating with a client, the device includes: The request receiving module is used to receive a video collection acquisition request sent by the client, wherein the video collection acquisition request includes a user identifier corresponding to the user. The category determination module is used to determine the user group category corresponding to the user based on the user identifier; The set matching module is used to determine the set of videos corresponding to the pre-edited tag information that matches the user group category as the target video set, wherein the video set is a set of videos obtained by mixing and cutting multiple videos to be edited; The video recommendation module is used to send the target video set to the client so that the client can play the target video set.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the method described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1-7.

Images