Video processing method and device based on fitness plan, equipment and storage medium

By generating fitness plans and monitoring user progress information, selecting suitable videos and adjusting content based on feedback, the problem of interruption of fitness exercises in the existing technology is solved, and the fitness effect and user experience are improved.

CN118984409APending Publication Date: 2024-11-19BEIJING XIAOTANG TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411100286.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-12
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

Existing video applications are difficult to ensure continuity during fitness exercises. Users need to stop frequently to find suitable videos, resulting in poor exercise results and distractions.

Method used

Generate fitness plans by obtaining user data, monitoring user progress information, obtain candidate video sets based on video characteristics, select suitable videos, and adjust video content based on user feedback to ensure the continuity of fitness activities.

Benefits of technology

It realizes the user's personalized fitness plan and continuously played fitness videos, improves the fitness effect and user experience, and avoids pauses and distractions in the middle of the fitness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118984409A_ABST
    Figure CN118984409A_ABST
Patent Text Reader

Abstract

The invention relates to a video processing method and device based on a fitness plan, equipment and a storage medium, and the method comprises the steps: obtaining user data, and generating a fitness plan which at least comprises fitness progress information and video features corresponding to the fitness progress information; after monitoring that the user is online, determining current progress information of the user, and obtaining a plurality of videos based on video features corresponding to the current progress information to obtain a candidate video set; selecting one or more candidate videos from the candidate video set to obtain a current fitness video of the user; monitoring feedback information of the current fitness video applied by the user; and adjusting the current fitness video or modifying the video feature corresponding to the current progress information based on the feedback information of the user. According to the embodiment of the invention, a perfect and personalized fitness plan and fitness videos or fitness video sets capable of being continuously played are provided for the user, and the fitness effect and application experience of the user are improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of video technology, and in particular to a video processing method, device, electronic equipment and storage medium based on a fitness plan. Background Art

[0002] With the improvement of living standards, people are generally concerned about their physical health. Whether it is the elderly, middle-aged people or young people, they generally have fitness methods that are suitable for them. For example, retired elderly people are very fond of square dancing, which not only exercises the body but also improves social interaction, making up for the lack of social activities of the elderly. However, the elderly who can participate in square dancing usually have good physical strength or extroverted personalities. For the elderly with poor physical strength or introverted personalities, this type of elderly people prefer some fitness activities that can be carried out in quiet and private places. For middle-aged and young people, some people prefer fitness venues, but for some people who are limited by time and energy, home is a better choice.

[0003] With the development of the Internet, applications (APPs) and video technology, more and more application content is presented to users in the form of videos. In particular, as people of all ages pay more and more attention to fitness and fitness is becoming more and more common, video content providers provide a variety of fitness content videos to meet this situation. For those elderly people who are not physically strong enough and introverted, or most middle-aged and young people, they can use video applications to exercise in any suitable place for them. Therefore, the rise of such video applications meets the fitness needs of many people.

[0004] Although current video applications provide a variety of fitness content for people's fitness needs, such as equipment application, fitness movements and training methods, yoga, meditation, Tai Chi, Tai Chi sword and other types of fitness activities, such as different durations of 2 minutes, 5 minutes, 30 minutes, or different exercise intensities and calorie consumption, thus meeting the needs of different users, it also brings a problem: in order to achieve better fitness results, users need to spend time to find videos that suit them from a large number of videos of many types, different durations, and different exercise intensities, and need to select multiple videos according to the duration of the video and their own exercise duration. In actual implementation, after users have spent a certain amount of time to find a suitable video and follow the instructor in the video to complete it, they still need to stop and continue to look for the next video, resulting in interruptions in fitness activities. This interruption not only interrupts the continuity of fitness, but also easily distracts users' attention when looking for videos, resulting in poor completion of fitness activities and failure to achieve the fitness results expected by users.

[0005] Therefore, there is an urgent need for a method to help users complete continuous fitness activities. Summary of the invention

[0006] In view of the technical problems existing in the prior art, the present invention proposes a video processing method, device, electronic device and storage medium based on a fitness plan, which can provide a fitness plan for users and ensure the continuity of users in fitness activities.

[0007] In order to solve the above technical problems, according to one aspect of the present invention, the present invention proposes a video processing method based on a fitness plan, comprising the following steps:

[0008] Acquire user data and generate a fitness plan, wherein the fitness plan at least includes fitness progress information and video features corresponding to the fitness progress information;

[0009] After monitoring that the user is online, determining the current progress information of the user, and acquiring multiple videos based on video features corresponding to the current progress information to obtain a candidate video set;

[0010] Select one or more candidate videos from the candidate video set to obtain a current fitness video of the user;

[0011] Monitor user feedback on the current fitness video; and

[0012] Adjust the current fitness video or modify the video features corresponding to the current progress information based on the user's feedback information.

[0013] According to another aspect of the present invention, the present invention also proposes a video processing device based on a fitness plan, the processing device comprising: a fitness plan module, a fitness video module and a feedback module, wherein the fitness plan module is configured to obtain user data and generate a fitness plan, wherein the fitness plan at least includes fitness progress information and video features corresponding to the progress information; the fitness video module is configured to determine the current progress information of the user after monitoring the user online, and obtain multiple videos based on the video features corresponding to the current progress information to obtain a candidate video set; select one or more candidate videos from the candidate video set to obtain the current fitness video of the user; the feedback module is connected to the fitness plan module and the fitness video module respectively, and is configured to monitor the feedback information of the user applying the current fitness video; wherein the fitness video module adjusts the current fitness video based on the user's feedback information; the fitness plan module modifies the video features corresponding to the current progress information based on the user's feedback information.

[0014] According to another aspect of the present invention, the present invention further proposes an electronic device, which includes a processor and a memory storing computer program instructions; when the electronic device executes the computer program instructions, the aforementioned video processing method based on the fitness plan is implemented.

[0015] According to another aspect of the present invention, the present invention further proposes a computer-readable storage medium, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned video processing method based on a fitness plan is implemented.

[0016] According to another aspect of the present invention, the present invention further proposes a computer program product, which includes computer program instructions, and when the computer program instructions are executed by a processor, the aforementioned video processing method based on a fitness plan is implemented.

[0017] The present invention provides users with a complete personalized fitness plan and a fitness video or fitness video collection that can be played continuously, so that users can continuously follow the fitness video to perform fitness activities, avoiding the interruption of fitness continuity and distraction of attention caused by users stopping to look for videos during fitness, thereby improving users' fitness effects and application experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The preferred embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, wherein:

[0019] Figure 1 is a flow chart of a video processing method based on a fitness plan according to an embodiment of the present invention;

[0020] Figure 2 is a flow chart of a method for generating a fitness plan according to an embodiment of the present invention;

[0021] Figure 3 is a processing flow chart of a method for extracting videos from a video set E to obtain a video set F according to an embodiment of the present invention;

[0022] Figure 4 is a schematic diagram of a video processing device based on a fitness program according to an embodiment of the present invention;

[0023] Figure 5 is a schematic diagram of the architecture of a video processing device based on a fitness program according to an embodiment of the present invention;

[0024] Figure 6 is a schematic diagram of a video processing device based on a client and a server according to an embodiment of the present invention; and

[0025] Figure 7FIG. 4 is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

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

[0027] In the following detailed description, reference may be made to the various specification drawings that are part of the present application and are used to illustrate specific embodiments of the present application. In the accompanying drawings, similar reference numerals describe substantially similar components in different figures. The various specific embodiments of the present application are described below in sufficient detail so that a person of ordinary skill in the art with relevant knowledge and skills in the art can implement the technical solutions of the present application. It should be understood that other embodiments may also be used or structural, logical or electrical changes may be made to the embodiments of the present application.

[0028] The present invention relates to a video processing method and device applied to a video platform, and the video platform includes but is not limited to a video application (APP) applied to a smart phone, a website provided in the form of a web page applied to a desktop computer, a tablet computer, a laptop computer, etc. The users of the video platform include video providing users and ordinary users who watch videos. In the following description, unless otherwise specified, the "user" in the present invention refers to an ordinary user who watches videos. After the user registers as a user on the video platform, in order to better serve the user, the video platform monitors and records the user's video playback data on the platform and stores it in a database to form a video playback record database. Each video playback data obtained is called a record, which includes the video identification, time and other related information played by the user. In order to avoid the user stopping to look for a video in the middle of fitness and causing interference with the continuity of fitness and distraction of attention, the present invention provides users with a complete personalized fitness plan and fitness videos or fitness video collections that can be played continuously.

[0029] Figure 1 1 is a flow chart of a video processing method based on a fitness plan according to an embodiment of the present invention. In this embodiment, the method comprises the following steps:

[0030] Step S1, acquiring user data and generating a fitness plan, wherein the fitness plan at least includes fitness progress information and video features corresponding to the progress information.

[0031] Step S2, monitor whether the user is online, and execute step S3 after monitoring that the user is online; if the user is not online, return to step S2.

[0032] Step S3, determining the current progress information of the user and the corresponding video features.

[0033] Step S4: acquiring multiple fitness videos based on the video features to obtain a candidate video set.

[0034] Step S5: Select one or more candidate videos from the candidate video set to obtain the current fitness video of the user.

[0035] Step S6, monitoring whether the user applies the current fitness video, if the user applies the current fitness video, executing step S7. If the user does not apply the current fitness video, returning to step S6.

[0036] Step S7, monitoring whether real-time feedback information of the user on the current fitness video is received, if real-time feedback information of the user on the current fitness video is received, executing step S8. If real-time feedback information of the user on the current fitness video is not received, returning to step S7.

[0037] Step S8, modifying the current fitness video and the video features corresponding to the current progress information based on the user's feedback information.

[0038] In step S1, the user data includes, for example, the user's personal information and historical video playback records. Figure 2 , Figure 2 1 is a flow chart of a method for generating a fitness plan according to an embodiment of the present invention, the method comprising the following steps:

[0039] Step S11, obtaining the video identifiers and playback features of a preset time period from the user's historical video playback records, wherein the playback features at least include the playback duration and the playback start time.

[0040] Step S12, obtaining the video features of each video from the video information library based on the video identifier.

[0041] Step S13, counting the playing characteristics of multiple videos in a preset time period.

[0042] Step S14, calculating the difference between the playback feature of each video and the corresponding video feature.

[0043] Step S15, determining the video preference characteristics and exercise preference characteristics of the user based on the statistical playback characteristics and the differences, wherein the video preference characteristics of the user include one or more video characteristics; and the exercise preference characteristics include one or more of exercise duration, exercise frequency and exercise intensity.

[0044] Step S16, obtaining user personal information, and determining fitness progress information based on the personal information.

[0045] Step S17, determining the video feature corresponding to the progress information based on the video preference feature and the exercise preference feature.

[0046] In the present invention, the user video playback record database stores a historical video playback record table for each user, and each record records the user's playback behavior on the video platform and the data of a corresponding video, including the recording time, video identification, and relevant information reflecting the video playback. Video identifications include, for example, video names, video identification (ID), or video providing user names, etc. Relevant information used to reflect video playback includes, for example, the playback start time, the playback end time, etc. According to the recording time, according to a preset time period, such as the same day, three days, one week, two weeks, one month or one year, the video identifications and corresponding playback information of all video playback records within the time are read in step S11.

[0047] In this embodiment, in order to facilitate understanding of the content of each video and for related business processing, the video platform sets multiple tags for each video in the video library, each tag corresponds to a video feature, and the video feature includes, for example, the following types of video features and corresponding feature information:

[0048] Fitness activity type: its specific characteristic information includes "aerobics", "belly slimming exercises", "body shape", "lazy exercises", "cervical spine exercises", "yoga", "Pilates", "whole body sweating" or "patting exercises", etc.

[0049] Exercise type: Its specific feature information includes "aerobic exercise", "anaerobic exercise", "strength training", etc., which can help users choose according to their exercise goals and preferences.

[0050] Action type: Its specific feature information includes "few actions", "moderate actions" or "many actions".

[0051] Video length: Its specific feature information includes "2 minutes", "5 minutes" or "30 minutes", etc.

[0052] Music type: Its specific feature information includes "popular songs", "classic old songs", "DJ dance music", "love songs", "folk songs", "red songs", "children's songs", etc.

[0053] Music rhythm: Its specific feature information includes "fast", "slow", "soothing", etc. Since fast-paced music can stimulate the user's enthusiasm for exercise, slow-paced music can help the user to do deep stretching or meditation. Therefore, the present invention also regards the background music rhythm in the fitness video as a video feature.

[0054] Interactivity: It is used to reflect whether the video contains interactive parts with the audience. Its specific feature information includes the coach's encouraging words, audience feedback, etc. Because different users have different interactive needs and preferences, the video feature of "interactivity" can help users find videos that better suit their interactive preferences.

[0055] Applicable groups: used to indicate which groups of people are suitable for watching the video. Specific characteristic information may include "beginners", "intermediate fitness enthusiasts", "advanced fitness enthusiasts", or specific groups such as "pregnant women", "elderly people", etc.

[0056] Safety Tips: This feature indicates whether the video contains tips for correct posture and avoiding injuries, which can help users exercise more safely.

[0057] Exercise intensity: used to reflect the exercise intensity of fitness activities in the video. Its specific feature information includes "low intensity", "medium intensity" and "high intensity".

[0058] Calorie consumption: used to reflect the calorie consumption of fitness activities in the video. Its specific characteristic information is such as "100 calories", "200 calories", etc.

[0059] Movement difficulty: used to reflect the difficulty of fitness movements in the video. Its specific feature information includes "entry level", "first-level difficulty", "second-level difficulty", "professional level", etc.

[0060] Targeted Part: Used to indicate the specific part of the body that the exercises in the video mainly target. Its specific feature information includes "abdomen", "legs", "back", etc.

[0061] Coach qualifications: used to introduce the professional background and qualifications of the coach in the video, and increase users' trust in the professionalism of the video content.

[0062] The feature information of the aforementioned various video features of each video is stored in the video information library in the form of tags, and corresponds to the video identifier one by one. Therefore, when a video identifier of a video is obtained, in step S12, the video information library is searched based on the video identifier, and the feature information of the video feature corresponding to it can be found.

[0063] In step S13, the playback characteristics of each video are counted based on the playback information of each video. The playback characteristics are, for example, "playback duration", and the specific value can be calculated from the playback start time and the playback end time. The playback characteristics can also include the number of playbacks, and the specific value is obtained by counting the number of the same video identifier within the time period.

[0064] In step S14, firstly, the difference between the playing time and the video time of each video is calculated, and based on the difference, the videos with a difference greater than a threshold are screened out. Then, according to the video features, the playing time and the difference with the video time of the videos with the same label content (feature information of the video features) are counted respectively. For example, there are n videos in the screened video set, among which for the video feature of "sports type", there are n1 "aerobic exercise" videos, n2 "anaerobic exercise" videos, and n3 "strength training" videos. The playing time of all the videos labeled "aerobic exercise" is added together to obtain the playing time T of the "aerobic exercise" videos. n1 , add up the duration of all videos labeled "aerobic exercise" to get the duration of the video "aerobic exercise" T tn1 , and calculate the difference D between the playback time and the video time n1 , where D n1 =T tn1 -T n1 Similarly, we can get the playback time T of the video labeled “anaerobic exercise” n2 , the difference between the playback time and the video time D n2 , the playback time of videos tagged with "strength training" is T n3 , the difference between the playback time and the video time D n3 In the same way, the play time and the difference between the play time and the video time of other video features are obtained respectively. In addition, the difference between the play time and the video time can also be replaced by the ratio of the play time to the video time. In addition, the ratio of the total play time of a video with a specific feature information to the total play time (total video time) of all videos played by the user can also be calculated as a preference factor.

[0065] Then in step S15, in an embodiment of determining the video preference feature and exercise preference feature of the user based on the statistical playback feature and the difference, the playback time of the video with the same feature information labeled as the same video feature is compared with the threshold value, so as to determine the user's preference for the feature information of the video feature. For example, for the three feature information of the video feature "exercise type", "anaerobic exercise", and "strength training", it is assumed that 2 / 3 of the video time is determined as the first threshold value, and 1 / 3 of the video time is determined as the second threshold value, and then the playback time of the video of each feature information is compared with the preference threshold value. When the playback time is greater than or equal to the first threshold value, the feature information is the user's favorite type. If the playback time is between the two threshold values, the feature information is the user's neutral type. If the playback time is less than the second threshold value, the feature information is the user's dislike type.

[0066] In another embodiment, the difference or ratio between the aforementioned playback duration and the video duration may be used as a preference factor, and the user's preference may be determined by a corresponding threshold.

[0067] Either the first two preference factors are combined to determine the user's preference, or the feature of "number of plays" is added as a preference factor to jointly determine the user's preference.

[0068] Or in another embodiment, when determining the user preference, it can be quantified into a numerical value, for example, the difference or ratio between the playback time and the video time is used as the quantified preference value.

[0069] The user's preferences are determined for other video features through the aforementioned method, and then the specific feature information of the video features that the user likes is counted, and the video preference features and exercise preference features of the user are determined according to the categories of the video features. In one embodiment, the user's video preference features include, for example, "aerobics", "body", "yoga" and the like in fitness activity types, and also include, for example, "aerobics" in sports types, "DJ dance music" in music types, and the like. The exercise preference features include the feature information related to exercise that the user likes determined from the video features, such as "medium intensity" in exercise intensity, "first-order difficulty" in action difficulty, and the like. The exercise preference features also include "exercise duration", "exercise frequency", "exercise preference time" and other features determined according to the time of each record in the user's historical video playback records, thereby obtaining a feature set that satisfies the user's preferences.

[0070] In step S16, the user personal information obtained includes, for example, the user's age, gender, position, physical data (such as weight, height, etc.) and other information filled in when the user registers, and then a fitness plan is determined based on the personal information, and the fitness plan includes fitness progress information. Wherein, when determining the fitness plan based on the personal information, in one embodiment, a fitness plan template adapted to the personal information is matched based on a preset fitness plan template. The fitness plan template is, for example, a fitness plan suitable for various groups of people determined based on the prior knowledge of experts, and the various groups of people can be determined by multiple dimensional information, such as age, gender, occupation, and weight. The fitness plan may include a total number of times or a total time span, such as 30 times, half a year, one month, three months, etc. The fitness plan may also include multiple stages divided by time nodes, for example, for a 6-month fitness plan, the same day of each month from the start date is used as a time node, and the entire fitness plan is divided into 6 stages; or the total number of times is divided by the number of times to obtain multiple stages, such as for a 30-time fitness plan, every 5 times is used as a segmentation point, and the entire fitness plan is divided into 6 stages. The above time nodes and the number of times are called progress nodes. Each stage further includes secondary time nodes and secondary progress nodes. The secondary time nodes and secondary progress nodes can be determined based on the user's exercise preference characteristics, such as "exercise frequency", "exercise preference time", etc.

[0071] The fitness plan may also include fitness indicator data corresponding to each progress node, such as the tasks to be completed and / or the goals to be achieved at each stage. The fitness indicators are, for example, the shortest exercise duration, the minimum calorie consumption, the intensity to be taken during exercise, etc., collectively referred to as progress information. By matching personal information and the applicable population dimension information of the template, a fitness plan template is determined for the user from a large number of fitness plan templates, and then the corresponding progress information is determined based on the exercise preference characteristics used. In another embodiment, a large model can be used to construct prompt information based on the user's personal information and exercise preference characteristics, and the fitness plan is output by the large model.

[0072] In step S17, the video features corresponding to the progress information are determined based on the video preference features and the exercise preference features. Specifically, based on the progress node information in the progress information of the fitness plan, the video features that meet the fitness indicators corresponding to the current progress node are selected from the feature set of user preferences, thereby forming a complete fitness plan. For example, based on the tasks and / or goals to be achieved corresponding to the progress nodes, the specific feature information of the corresponding exercise-related video features is determined according to the user's exercise preference features, such as the specific feature information of the exercise intensity, the difficulty of the action, the calorie consumption, etc., and the specific video feature information corresponding to the specific video preference features. Taking a 6-month fitness plan as an example, for each exercise stage, the corresponding video features include specific feature information of the video features that reflect the user's video preferences, such as "aerobic exercise" in the sports type, "Pilates" in the fitness activity type, "classic old songs" in the music type, etc. The specific feature information of these video features is the video preference features of the user obtained above. The video features corresponding to each exercise stage also include video features that reflect the user's exercise preferences, and match the progress information of the exercise stage, such as the "video duration" in the video features corresponding to the "minimum exercise duration" in the progress information, the "calorie consumption" in the video features corresponding to the "minimum calorie consumption" in the progress information, and the "exercise intensity" in the video features corresponding to the "exercise intensity" in the progress information.

[0073] After a complete fitness plan is generated for a user, it is stored in a database as a kind of personal information of the user, and the user can view it through a personal account after logging into the video platform. In another embodiment, the user may also be allowed to modify it. In addition, the video platform records the user's application data to the fitness plan, such as recording the user's playback information of a fitness video or video set generated based on the fitness plan, such as the playback time, the played fitness video, etc., and based on the progress information in the fitness plan, determines the stage at which the user's playback behavior of the fitness video occurs during the recording, thereby marking the progress information in the fitness plan to obtain the progress marking information.

[0074] In order to facilitate the user's application of the fitness plan, the present invention monitors the user's online status. When the user is detected to be online, in step S3, the user's current progress information is determined by querying the progress mark information of the fitness plan, and a corresponding video feature set is obtained.

[0075] In step S4, when acquiring multiple fitness videos based on the video features to obtain a candidate video set, there can be multiple methods for obtaining the candidate video set, for example, determining a first video feature corresponding to a video preference feature and a second video feature corresponding to an exercise preference feature in a video feature set corresponding to current progress information; then matching multiple fitness videos from a video database according to the first video feature to obtain a first video set; and then matching multiple fitness videos from the first video set according to the second video feature to obtain a candidate video set.

[0076] For example, for the video feature "fitness activity type", the feature information preferred by user A is "aerobics", "cervical spine exercises" and "clapping exercises", and the respective preference scores calculated by quantifying the preferences are 1, 0.8 and 0.6 respectively; for the video feature "music type", the feature information preferred by user A is "popular songs", "love songs" and "folk songs", and the respective quantified preference scores are 1, 0.7 and 0.3 respectively. Then, a complete video and / or video clip list whose "fitness activity type" meets "aerobics", "cervical spine exercises" and "clapping exercises" and whose "music type" meets "popular songs", "love songs" and "folk songs" is screened out from the video library, and stored together with the preference scores as the first video set.

[0077] It is also known that the exercise preference characteristics of user A are: exercise duration equal to 30 minutes, exercise intensity is "medium intensity", and action type is "moderate action". Therefore, videos with a video duration of 30 minutes, exercise intensity of "medium intensity", and action type of "moderate action" are selected from the first video set as the candidate video set.

[0078] In another embodiment, after matching multiple fitness videos from the video database according to the first video feature to obtain the first video set, the videos in the first video set are videos that meet the user's interests. When matching multiple fitness videos from the first video set according to the second video feature to obtain the candidate video set, the matching degree of the second video feature and each fitness video in the first video set is calculated, and the fitness videos with a matching degree greater than or equal to the threshold are determined as the first video group, and the fitness videos with a matching degree less than the threshold are determined as the second video group. The first video group and the second video group constitute the candidate video set. From the aforementioned processing, it can be seen that the videos in the first video group are highly consistent with the user's exercise preferences, and the videos in the second video group are less consistent with the user's exercise preferences. For example, when filtering videos from the first video set according to the video duration of 30 minutes, the exercise intensity of "medium intensity" and the action type of "moderate action", it is assumed that the threshold is 1, that is, the videos in the first video set must completely match the aforementioned three video features, and the completely matching videos are combined into the first video group, and the rest are the second video group. That is to say, videos with a duration of less than or equal to 30 minutes, a workout intensity of "medium intensity" and an action type of "moderate action" are classified into the first video group, while videos with a duration of less than 30 minutes, a workout intensity of "low intensity" or "high intensity", and action types of "many actions" and "few actions" are classified into the second video group.

[0079] When the first video feature is defined as corresponding to the exercise preference feature and the second video feature is defined as corresponding to the video preference feature, when a plurality of fitness videos are matched from the video database according to the first video feature to obtain the first video set, the matching degree between the first video feature and each fitness video in the video database is calculated, and the fitness videos with a matching degree greater than or equal to the threshold are determined as the third video group, and the fitness videos with a matching degree less than the threshold are determined as the fourth video group, and the third video group and the fourth video group constitute the first video set. That is, firstly, according to the user's exercise preference classification, the video database determines the videos that are highly consistent with the exercise preference and the videos with low consistency. At this time, two types of videos can be selected according to the preset quantity. Then, according to the second video feature, the third video group and the fourth video group are matched to the corresponding plurality of videos, and the plurality of videos matched from the third video group are determined as the first video group, and the plurality of videos matched from the fourth video group are determined as the second video group, and the first video group and the second video group constitute the candidate video set. That is, according to the user's video interest, videos that match the video interest are screened out from the two categories respectively, thereby obtaining a first video group that matches the user's video preference and exercise preference and a second video group that matches the user's video preference but has a low match with the exercise preference.

[0080] In another further embodiment, the present invention monitors the user's feedback on the fitness video, and determines negative video features based on the feedback information. The negative video features, for example, reflect the coach that the user dislikes, the type of music that the user dislikes, the inappropriate exercise intensity, etc. Each time the user's feedback information is obtained, the corresponding video features are extracted, which are referred to as negative video features in the present invention, and the negative video features are recorded and marked. Therefore, in this embodiment, after obtaining a video set according to the first video feature and the second video feature in accordance with the various embodiments described above, which is referred to as the first candidate video set in this embodiment, negative video features are further determined from the video feature set corresponding to the fitness progress information, for example, video features marked as negative are read according to the marked content, or negative video features are read from a video set marked or classified as negative video features, and videos in the first candidate video set are filtered based on the negative video features, and the filtered videos constitute the candidate video set. Then step S5 is performed.

[0081] In step S5, multiple methods may be used to select one or more candidate videos from the candidate video set to obtain the current fitness video of the user, such as random rules, statistical rules, business rules and mixed rules, wherein the mixed rule is to use at least two of the random rules, statistical rules and business rules in combination. Specifically, firstly, the duration T of the fitness video to be generated is determined based on the shortest duration of exercise corresponding to the current progress information, and corresponding to the random rule, multiple videos that meet the duration T are randomly selected from the candidate video set.

[0082] For statistical rules, for example, the matching degree between each video and the user preference is counted, and multiple videos satisfying the duration T are randomly selected from the candidate video set in the order of matching degree from high to low.

[0083] For business rules, the business rules are, for example, some conditions determined by business features extracted from certain businesses of the video platform. For example, the business is a membership business. For video providing users, based on payment conditions, paid video providing users are determined as members, and non-paid video providing users are determined as non-members. For members, they are divided into multiple levels according to different payments, so videos of video providing users with high levels are preferentially selected from the candidate video set.

[0084] When the candidate video set includes a first video group of a high-weight group and a second video group of a low-weight group, the specific statistical processing process includes:

[0085] First, the preference probability f(x) of each video in the candidate video set is calculated according to the following formula 1-1:

[0086]

[0087] Where x represents a single video, H is the first video group in the candidate video set as a high-weight group, and L is the second video group in the candidate video set as a low-weight group; μ (μ∈(0.5,1]) is the probability of the set high-weight group appearing, and α and β are the preference scores of the video features, respectively.

[0088] Then, the preference probability f(x) of each video is compared with the screening threshold probability. When the preference probability f(x) of a video is greater than the threshold probability, the video is selected, thereby obtaining a preliminary video set, which is recorded as video set E. Therefore, the current video set E includes a certain number of videos in the first video group and a certain number of candidate videos in the second video group.

[0089] Then, multiple videos are randomly selected from the video set E as the video set constituting the fitness video, which is recorded as the video set F, so that the cumulative video duration of the video set F is approximately equal to the duration T of the fitness video to be generated. Figure 3 , Figure 3 1 is a processing flow chart of an extraction method for extracting videos from a video set E to obtain a video set F according to an embodiment of the present invention. The video to be extracted is recorded as the jth video, j=1,2,3,...,n, and n is the total number of videos in the video set E. In the first processing, j=1.

[0090] Step S511: randomly select a video from the video set E as the jth video.

[0091] Step S512: Calculate the difference L between the cumulative duration of the video set F and the duration T of the fitness video to be generated according to Formula 1-2.

[0092]

[0093] Where |F| is the number of videos in the video set F, t j is the duration of the jth video extracted from the video set E, T is the duration of the fitness video to be generated, t i is the duration of the i-th video in the video set F.

[0094] That is, when there is no video in the video set F, the difference L is equal to the difference between the duration of the j-th video and the duration T of the fitness video to be generated. When there are i videos in the video set F, the difference L is equal to the difference between the sum of the durations of the i videos and the sum of the durations of the j-th video and the duration T of the fitness video to be generated.

[0095] Step S513, compare whether the difference L is less than the set threshold -delta. When the difference L is less than the set threshold -delta, it means that the duration obtained when adding the jth video to the video set F is much less than the duration T of the fitness video to be generated, so in step S514, the current jth video is extracted into the video set F. If the difference L is greater than the set threshold -delta, then in step S518, determine whether the difference L is between -delta and delta. If the difference L is between -delta and delta, it means that after adding the current jth video, the cumulative duration of the videos in the video set F basically reaches the duration T of the fitness video to be generated, then in step S519, the current jth video is extracted into the video set F, and the extraction process ends. If the difference L is not between -delta and delta, that is, the difference L is greater than delta, it means that after adding the current j-th video, the cumulative duration of the videos in the video set F exceeds the duration T of the fitness video to be generated by too much, so it will not be added to the video set F, and step S515 will be executed.

[0096] Step S515, set j=j+1.

[0097] Step S516, determine whether j is equal to n. If j is equal to n, it means that there are no videos to be extracted in the video set E, and the number of videos in the video set E is too small. Then, in step S517, a certain number of videos are screened from the candidate video set and added to the video set E, and the process returns to step S511. For example, the screening threshold probability is reduced to screen out a certain number of more videos. If j is less than n, it means that there are still unextracted videos in the video set E, and the process returns to step S511. When randomly selecting videos from the video set E, the present invention randomly selects without replacement, thereby avoiding the problem of repeated selection of videos.

[0098] The video set F obtained through the aforementioned processing process includes all videos constituting the fitness video. In one embodiment, all videos in the video set F can be combined into one video as the current fitness video of the user, or all videos in the video set F can be combined into a video set played sequentially as the current fitness video of the user.

[0099] In addition, when the multiple candidate videos are combined into one video, or when the multiple candidate videos are combined into one video set to be played sequentially, the multiple candidate videos are sorted according to a preset sorting strategy rule.

[0100] The preset sorting strategy rules include a combination of one or more of the following strategies: random sorting, sorting by video duration from long to short or from short to long, sorting by configuration scheme of corresponding feature information of one or more preset video features, and sorting by configuration scheme based on business features.

[0101] There are many different configuration schemes according to the configuration scheme of the feature information of one or more preset video features, for example:

[0102] Configuration scheme 1: The order of configuration based on the characteristic information of exercise intensity is: low-medium-medium-low-high-low-medium-medium-low;

[0103] Configuration scheme 2: Sort videos based on the user's preference probability f(x) from high to low;

[0104] Configuration scheme three: The order of configuration based on feature information of action types is: few actions - moderate actions - many actions - few actions - moderate actions - moderate actions - moderate actions - few actions - many actions - moderate actions;

[0105] Configuration scheme 4: The order of configuration based on the characteristic information of music rhythm is: slow-fast-slow-fast-fast-slow-slow-fast-slow-slow;

[0106] Configuration scheme 5: sort the videos in the current video set F according to the user level (i.e., the user level of the video provided) in descending order;

[0107] Configuration scheme six: Sort the videos in the current video set F by using the service features as the first priority level. For videos of the same priority level, random sorting or any one of the above configuration schemes one to four is used.

[0108] In step S7, when monitoring the feedback information of the user applying the current fitness video, it is monitored whether the user's screen-based operation, such as screen click, screen sliding, etc., is received during the playback of the current fitness video. Among them, during the playback of the current fitness video, an interactive interactive entrance is provided on the playback interface, such as a heart-shaped button for indicating that the video is liked, a thumbs-down button for indicating that the video is not liked, or a button or label, page, etc. for entering the feedback window. The user's feedback information can also be provided in the form of voice, for example, during the playback of the current fitness video, the microphone of the user device is turned on to listen to the user's voice, and when the user's voice information is extracted, the operation information given by the user is identified therefrom. Of course, the user can also feedback relevant information by using gesture operations, for example, during the playback of the current fitness video, the camera of the user device is turned on to capture the user's gestures, such as the upward or downward or left or right waving action when the user stops the action, so as to identify the user's instruction to skip the current video. In addition, the posture of the user device can also be monitored, and the specific user instruction can be identified according to the change of the posture. For example, when the user device tilts to one side once or several times, the action is identified as a user instruction to skip the current video.

[0109] Based on the aforementioned various types of operations, corresponding operation information can be obtained, such as "I don't like teacher A", "I don't like song B", "I don't like background C", "I'm tired of listening to song D", "I'm tired of dancing action E", "Video F has poor image or sound quality", "Action G is not suitable", "Video H has inappropriate intensity", "I don't like the command for video I", etc.

[0110] When the user's operation information is identified, negative video features are extracted from the operation information. For example, when the operation information is "the intensity of video H is not appropriate", the video feature can be extracted as "exercise intensity". Based on the label of the current video, the specific feature information of "exercise intensity" can be obtained, such as "high intensity". Based on the user's operation information of "the intensity of video H is not appropriate", the "high intensity" is determined as a negative video feature. For another example, the video feature "coach qualifications" includes the personal information of the coach in the video. Most of the videos in the video library have corresponding "coach qualifications" tags. When the user's operation information is "I don't like teacher A", the specific feature information "coach A" of the video feature "coach qualifications" is used as a negative video feature.

[0111] When negative video features are extracted, the current fitness video is adjusted in step S8. In one embodiment, when adjusting the fitness video, one or more videos are determined from the candidate videos to replace the videos in the current fitness video that include negative video features. For example, when the received operation information is "inappropriate video intensity" and "high intensity" is determined to be a negative video feature, all videos including the "high intensity" label are deleted from the current fitness video set, and one or more videos with specific feature information of "exercise intensity" of "low intensity" or "medium intensity" are determined from the candidate videos (such as the first video group and the second video group mentioned above) and added to the current fitness video set. The selection process is as follows: Figure 3 As shown, no further description is given here.

[0112] In another embodiment, when adjusting a fitness video, the negative video features and the corresponding video features of the current fitness video are analyzed, and the adjustment method is determined based on the analysis results. For example, for the specific feature information "high intensity" of the aforementioned video feature "exercise intensity", the number of videos in the current fitness video set and the number of videos or video durations labeled "high intensity" are queried. When the number of videos or video durations with the current label "high intensity" accounts for a higher proportion than a threshold value (such as one-third), the number of videos or durations that need to be changed are calculated, and then the videos that include the number of videos or durations that need to be changed with the "high intensity" label are deleted from the current fitness video set, and then the videos whose specific feature information of "exercise intensity" corresponding to the changed number or duration is "low intensity" or "medium intensity" are determined from the candidate videos (such as the aforementioned first video group and second video group) and added to the current fitness video set. Then, the user is continuously monitored to see if there is any feedback information on the negative video feature "high intensity". If so, the video with this feature information is continuously reduced. If not, it means that the overall exercise intensity of the current video is suitable for the user, and the adjustment of the negative video feature is terminated. Video features that may be processed similarly may include, for example, “video duration”, “sports type”, “fitness activity type”, “music rhythm”, “movement difficulty”, and the like.

[0113] This embodiment can not only generate a continuously played video or video collection for the user, but also make adjustments in real time according to the user's feedback, so as to better meet the user's preferences and needs.

[0114] Further, after adjusting the current fitness video set, modify the video features corresponding to the current progress information in the fitness plan. Wherein, when modifying the video features corresponding to the current progress information, in one embodiment, delete all the negative video features extracted above. In another scheme, analyze each negative video feature, and determine whether to delete or modify it based on the analysis result. For example, when the negative video feature is the specific feature information "Coach A" of "Coach Qualifications", delete the specific feature information "Coach A" of the video feature "Coach Qualifications" from the video feature set corresponding to the current progress information; or delete and add the video feature set to be filtered in the video feature set corresponding to the current progress information, and add "Coach A" to the filtered video feature set, or mark the video features in the video feature set, and mark "Coach A" as the feature information to be filtered.

[0115] It may further include: querying historical feedback information, if the historical feedback information includes the negative video features (such as the specific "coach A") and the number of times is more than a threshold (such as 3 times, 5 times, etc.), deleting and adding the video feature set to be filtered in the video feature set corresponding to each progress information, and adding "coach A" to the filtered video feature set, or marking the video features in the video feature set, marking "coach A" as the feature information to be filtered. Therefore, the next time the user logs in to the video platform, when generating a fitness video for the user, the video including the negative video features is filtered out, thereby better meeting the user's preferences and reducing the impact of the user's negative feedback on the fitness activities during the fitness video application process.

[0116] Figure 4 1 is a schematic diagram of a video processing device based on a fitness plan according to an embodiment of the present invention, wherein the video processing device based on a fitness plan in this embodiment comprises a fitness plan module 10, a fitness video module 20 and a feedback module 30. The fitness plan module 10 is connected to a database of a video platform and is configured to obtain user data from the database and generate a fitness plan, wherein the fitness plan at least comprises fitness progress information and video features corresponding to the progress information.

[0117] The fitness video module 20 is configured to read the user's fitness plan from the database after monitoring the user online, thereby determining the user's current progress information, and obtain multiple fitness videos based on the video features corresponding to the current progress information to obtain a candidate video set; select one or more candidate videos from the candidate video set to obtain the user's current fitness video.

[0118] The feedback module 30 is connected to the fitness plan module 10 and the fitness video module 20 respectively, and is configured to monitor the feedback information of the user applying the current fitness video, store the feedback information in a database, and send it to the fitness plan module 10 and the fitness video module 20 at the same time. The fitness video module 20 adjusts the current fitness video based on the user's feedback information, and the fitness plan module 10 modifies the video features corresponding to the current progress information based on the user's feedback information.

[0119] The video processing device of this embodiment can provide users with a complete personalized fitness plan and continuously played fitness videos or fitness video collections, avoiding the interruption of fitness continuity and distraction of attention caused by users stopping to look for videos during fitness.

[0120] Figure 5 : is an architectural schematic diagram of a video processing device based on a fitness plan according to another embodiment of the present invention. In this embodiment, the video processing device includes a client and a server. The client device is located in a user device, which can be a separate client application, or a plug-in or applet of a video platform. The user device is, for example, a mobile terminal (such as a smart phone) 100 or a computer 102. The server is located in a server 200, and the server 200 is connected to a database 300. The client and the server use the network unit in the device to achieve network communication, and then send various requests and various data to each other.

[0121] See also Figure 6 , Figure 61 is a schematic diagram of a video processing device based on a client and a server according to an embodiment of the present invention. In this embodiment, the client includes a personal information unit 110, a video playback unit 120, a feedback monitoring unit 130, and a client communication unit 140. The server includes a fitness plan unit 210, a user online monitoring unit 220, a fitness video unit 230, a feedback information processing unit 240, a server communication unit, and a database 300. The fitness plan unit 210 of the server obtains user data from the database 300 to generate a personalized fitness plan for each user. The user online monitoring unit 220 monitors the user status and notifies the fitness video unit 230 after monitoring that the user is online. The fitness video unit 230 reads the user's fitness plan from the database 300, thereby determining the user's current progress information, and generates the user's current fitness video based on the video features corresponding to the current progress information. In one embodiment, the current fitness video or its information is sent to the client via the server communication unit 250. After receiving the current fitness video or its information, the client communication unit 140 of the client sends it to the personal information unit 110, displays the current fitness video or its information in the corresponding section and displays the play button provided by the video playback unit 120. When the user applies the current fitness video, click the play button. When the video playback unit 120 receives the play button information, when the current fitness video displayed in the personal information unit 110, that is, when the client has received the complete current fitness video, the current fitness video is played in the user device. When the current fitness video is a video, it is played from the beginning. When the current fitness video is a video set, it is played in sequence according to the preset order. If the client receives only the information of the current fitness video, such as the video cover, video logo, etc., the video playback unit 120 requests the corresponding video from the server through the client communication unit 140. After receiving the video request, the server communication unit 250 of the server sends its request information to the fitness video unit 230, and the fitness video unit 230 sends the data of the current fitness video to the client, so that it is played online in real time in the user device.

[0122] The feedback monitoring unit 130 monitors the user feedback information during the current fitness video playback process in real time. After monitoring the user feedback information, the user's operation information is obtained therefrom, and the operation information is sent to the server through the client communication unit 140. The feedback information processing unit 240 of the server receives the user's operation information, extracts negative video features therefrom, stores them in the database 300, and sends them to the fitness plan unit 210 and the fitness video unit 230 at the same time. The current fitness video is adjusted by the fitness video unit 230, and the adjusted information or video is sent to the client. The video features corresponding to the current progress information are modified by the fitness plan unit 210. The specific processing process refers to the aforementioned description of the method, which will not be repeated here.

[0123] The present invention also provides an electronic device, comprising a processor and a memory storing computer program instructions; when the processor executes the computer program instructions, the video processing method based on the fitness plan in any of the above embodiments is implemented. The electronic device in the present invention can be Figure 5 The user equipment in the embodiment, such as the mobile terminal 100 or the computer 102, may also be Figure 5 Server 200 in. Figure 7 FIG. 1 is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Figure 7 As shown, the electronic device may include a processor 601 and a memory 602 storing computer program instructions.

[0124] Specifically, the processor 601 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiment of the present invention.

[0125] The memory 602 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 602 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 602 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 602 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 602 is a non-volatile solid-state memory.

[0126] The memory 602 may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Therefore, generally, the memory 602 includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described in any of the video processing methods based on fitness programs of the present invention.

[0127] The processor 601 reads and executes the computer program instructions stored in the memory 602 to implement any one of the video processing methods based on the fitness plan in the above embodiments.

[0128] In one example, the electronic device may further include a communication interface 603 and a bus 610. Figure 7 As shown, the processor 601, the memory 602, and the communication interface 603 are connected through the bus 610 and complete the communication between each other. The communication interface 603 is mainly used to realize the communication between each module, device, unit and / or device in the embodiment of the present invention. The bus 610 includes hardware, software or both, and couples the components of the online data flow billing device to each other. For example and not limitation, the bus may include an accelerated graphics port (AGP) or other graphics bus, an enhanced industrial standard architecture (EISA) bus, a front-end bus (FSB), a hypertransport (HT) interconnect, an industrial standard architecture (ISA) bus, an infinite bandwidth interconnect, a low pin count (LPC) bus, a memory bus, a micro channel architecture (MCA) bus, a peripheral component interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a serial advanced technology attachment (SATA) bus, a video electronics standard association local (VLB) bus or other suitable buses or a combination of two or more of these. Where appropriate, the bus 610 may include one or more buses. Although the embodiments of the present invention describe and illustrate specific buses, the present invention considers any suitable bus or interconnect.

[0129] In addition, in combination with the video processing method based on the fitness plan in the above embodiment, the embodiment of the present invention can provide a computer-readable storage medium and a computer program product. The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by the processor, any one of the video processing methods based on the fitness plan in the above embodiment is implemented. The computer-readable storage medium is, for example, the various memories mentioned above. The computer program product described in the present invention can be a software package or code segment that can be downloaded via a computer network such as the Internet, an intranet, etc.

[0130] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0131] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Computer-readable storage medium" may include any medium capable of storing or transmitting information, may include a non-transient computer-readable storage medium, such as including an electronic circuit, a semiconductor memory device, a ROM, a flash memory, an erasable ROM (EROM), a floppy disk, a CD-ROM, an optical disk, a hard disk, an optical fiber medium, may also include a radio frequency (RF) link, etc.

[0132] It should also be noted that the exemplary embodiments mentioned in the present invention describe some methods or systems based on a series of steps or devices. However, the present invention is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiments, or in a different order from the embodiments, or several steps can be performed simultaneously.

[0133] The above embodiments are only used to illustrate the present invention, but not to limit the present invention. Ordinary technicians in the relevant technical field can make various changes and modifications without departing from the scope of the present invention. Therefore, all equivalent technical solutions should also fall within the scope of the present invention.

Claims

1. A video processing method based on a fitness program, characterized in that: include: Acquire user data and generate a fitness plan, wherein the fitness plan at least includes fitness progress information and video features corresponding to the fitness progress information; After monitoring that the user is online, determining the current progress information of the user, and acquiring multiple videos based on video features corresponding to the current progress information to obtain a candidate video set; Select one or more candidate videos from the candidate video set to obtain a current fitness video of the user; Monitor the feedback information of users applying the current fitness video; as well as Adjust the current fitness video or modify the video features corresponding to the current progress information based on the user's feedback information.

2. The method according to claim 1, characterized in that When selecting one or more candidate videos to obtain the current fitness video of the user, when multiple candidate videos are selected, the multiple candidate videos are combined into one video as the current fitness video of the user, or the multiple candidate videos are combined into a video set played sequentially as the current fitness video of the user.

3. The method according to claim 2, characterized in that When the multiple candidate videos are combined into one video, or when the multiple candidate videos are combined into one video set to be played sequentially, the multiple candidate videos are sorted according to a preset sorting strategy rule.

4. The method according to claim 3, characterized in that The preset sorting strategy rules include a combination of one or more of the following strategies: random sorting, sorting by video duration from long to short or from short to long, sorting by configuration scheme of corresponding feature information of one or more preset video features, and sorting by configuration scheme based on business features.

5. The method according to claim 1, characterized in that The steps to obtain user data and generate a fitness plan include: Obtaining the historical video playback records of the user within a preset time period, and obtaining a video identifier and playback information from each video historical playback record, wherein the playback information at least includes a playback start time and an end time; Obtaining video features of each video from a video information library based on the video identifier; Calculate the playback feature of each video based on the playback information of each video; Counting the play characteristics of the plurality of videos in the preset time period and calculating the difference between the play characteristics of each video and the corresponding video characteristics, and determining the video preference characteristics and the exercise preference characteristics of the user based on the counted play characteristics and the difference; and Obtain user personal information, determine a fitness plan based on the personal information, the fitness plan includes fitness progress information, and determine video features corresponding to the fitness progress information based on video preference features and exercise preference features to form a video feature set.

6. The method according to claim 5, characterized in that: The fitness progress information includes one or more of a progress node, a training duration, and a fitness index corresponding to each progress node.

7. The method according to claim 6, characterized in that: The step of acquiring multiple videos based on the video features corresponding to the current progress information to obtain a candidate video set includes: Determine, from the video feature set corresponding to the fitness progress information, a first video feature corresponding to the user's video preference feature / exercise preference feature and a second video feature corresponding to the user's exercise preference feature / video preference feature; According to the first video feature, multiple videos are matched from a video database to obtain a first video set; and According to the second video feature, multiple videos are matched from the first video set to obtain a candidate video set.

8. The method according to claim 7, characterized in that: When the first video feature corresponds to the user's video preference feature and the second video feature corresponds to the exercise preference feature, when matching multiple videos from the first video set according to the second video feature to obtain a candidate video set, calculating the matching degree between the second video feature and each video in the first video set, determining videos with a matching degree greater than or equal to a threshold as a first video group, and determining videos with a matching degree less than a threshold as a second video group, the first video group and the second video group constitute a candidate video set; Correspondingly, when selecting one or more candidate videos from the candidate video set to obtain the user's current fitness video, a first number of candidate videos are selected from the first video group, and a second number of candidate videos are selected from the second video group, and the first number of candidate videos and the second number of candidate videos constitute the current fitness video.

9. The method according to claim 7, characterized in that: When the first video feature corresponds to the user's exercise preference feature and the second video feature corresponds to the video preference feature, when a plurality of fitness videos are matched from a video database according to the first video feature to obtain a first video set, a matching degree between the first video feature and each video in the video database is calculated, and videos with a matching degree greater than or equal to a threshold are determined as a third video group, and videos with a matching degree less than a threshold are determined as a fourth video group, and the third video group and the fourth video group constitute a first video set; When a plurality of videos are matched from the first video set according to the second video feature to obtain a candidate video set, a plurality of corresponding videos are matched from the third video group and the fourth video group according to the second video feature, the plurality of videos matched from the third video group are determined as a first video group, and the plurality of videos matched from the fourth video group are determined as a second video group, and the first video group and the second video group constitute a candidate video set; Correspondingly, when selecting one or more candidate videos from the candidate video set to obtain the user's current fitness video, a first number of candidate videos are selected from the first video group, and a second number of candidate videos are selected from the second video group, and the first number of candidate videos and the second number of candidate videos constitute the current fitness video.

10. The method according to any one of claims 7 to 9, characterized in that: When a plurality of videos are matched from the first video set according to the second video feature, the plurality of videos constitute a first candidate video set, further comprising: determining negative video features from a set of video features corresponding to the fitness progress information; and The videos in the first candidate video set are filtered based on the negative video features, and the filtered videos constitute a candidate video set.

11. The method according to claim 1, characterized in that: The steps of monitoring the feedback information of the user applying the current fitness video include: Receiving operation information input by the user based on screen operation, voice operation, gesture operation or terminal posture operation; and Extracting corresponding negative video features from the operation information; Correspondingly, the step of modifying the video feature corresponding to the current progress information based on the user's feedback information includes one or more of the following operations: Filter out the extracted negative video features from the video features corresponding to the current progress information; Adding the negative video feature to the video feature corresponding to the current progress information and marking it; The negative video features extracted from the history are queried, and when the current negative video features meet the conditions, the negative video features corresponding to each progress information are deleted from the fitness plan.

12. A video processing device based on a fitness program, characterized in that: include: a fitness plan module configured to acquire user data and generate a fitness plan, wherein the fitness plan includes at least fitness progress information and video features corresponding to the progress information; The fitness video module is configured to determine the current progress information of the user after monitoring the user online, and obtain multiple videos based on the video features corresponding to the current progress information to obtain a candidate video set; select one or more candidate videos from the candidate video set to obtain the current fitness video of the user; as well as A feedback module, which is connected to the fitness plan module and the fitness video module respectively, and is configured to monitor the feedback information of the user applying the current fitness video; wherein the fitness video module adjusts the current fitness video based on the user's feedback information; The fitness plan module modifies video features corresponding to current progress information based on user feedback information.

13. An electronic device, characterized in that: The electronic device comprises a processor and a memory storing computer program instructions; when the electronic device executes the computer program instructions, the video processing method based on the fitness plan as described in any one of claims 1 to 11 is implemented.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the video processing method based on a fitness plan according to any one of claims 1 to 11 is implemented.

15. A computer program product, characterized in that The device comprises computer program instructions, which, when executed by a processor, implement the video processing method based on a fitness program as claimed in any one of claims 1 to 11.