Video playback method, apparatus, device, and storage medium

Abstract

The embodiments of the present disclosure disclose a video playing method, an apparatus, a device and a storage medium, which includes: detecting that a preset application is switched from a non-background running state to a background running state during playing a first video, continuing to play the first video in the background by the preset application, and determining a second video from a second video candidate set according to historical operation information of a current user for the preset application and a preset background video recommendation model, and playing the second video in the background by the preset application when detecting a preset video switching event.

Description

Technical Field

[0001] [Cross - Reference to Related Applications] This application claims the priority of a Chinese patent application filed with the State Intellectual Property Office of China on August 27, 2021, with the application number 202110996241.6, and all the contents of this application are incorporated herein by reference.

[0002] [Technical Field] Embodiments of the present disclosure relate to the field of Internet technology, for example, to video playback methods, devices, equipment, and storage media.

Background Art

[0003] With the rapid development of information technology and the popularization of mobile Internet, the network information that applications can provide to users is increasing, and among them, the video recommendation function of applications is favored by more and more users.

[0004] Currently, applications generally recommend videos to users in the way of pushing video streams. There may be situations where users cannot continue to watch the video screen during video viewing, such as the terminal screen turning off, switching to other applications, or switching to the terminal's desktop. In such cases, the push effect of the related video stream is not ideal and needs to be improved.

Summary of the Invention

Problems to be Solved by the Invention

[0005] Embodiments of the present disclosure provide a video playback method, device, storage media, and equipment that can optimize the conventional video playback solution.

Means for Solving the Problems

[0006] According to a first aspect, embodiments of the present disclosure provide a video playback method, and this method is Detecting that a preset application for playing a video stream is switched from a non-background running state to a background running state during the playback of a first video; Continuing to play the first video in the background by the preset application, and determining a second video from a set of second video candidates based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model, wherein the preset background video recommendation model is used to determine the background playback video to be recommended, the set of second video candidates is used to select a video therefrom during the background playback mode, the set of second video candidates is different from the set of first video candidates, and the set of first video candidates is used to select a video therefrom during the non-background playback mode; When a preset video switching event is detected, playing the second video in the background by the preset application.

[0007] According to a second aspect, an embodiment of the present disclosure provides a video playback device, which device includes: A state detection module arranged to detect whether a preset application for playing a video stream is switched from a non-background running state to a background running state during the playback of a first video; After detecting that the preset application is switched from a non-background running state to a background running state during the playback of the first video, the preset application continues to play the first video in the background, and based on the historical operation information of the current user with respect to the preset application and the preset background video recommendation model, a video determination module is arranged to determine a second video from a set of second video candidates. The preset background video recommendation model is used to determine the background playback video to be recommended, the set of second video candidates is used to select a video therefrom during the background playback mode, the set of second video candidates is different from the set of first video candidates, and the set of first video candidates is used to select a video therefrom during the non-background playback mode. The video determination module, A playback switching module arranged to play the second video in the background by the preset application when a preset video switching event is detected.

[0008] According to a third aspect, an embodiment of the present disclosure provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, a video playback method according to an embodiment of the present disclosure is realized.

[0009] According to a fourth aspect, an embodiment of the present disclosure provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, a video playback method according to an embodiment of the present disclosure is realized.

Brief Description of the Drawings

[0010]

Figure 1

Figure 2

Figure 3

Figure 4

DETAILED DESCRIPTION OF THE INVENTION

[0011] The following describes embodiments of the present disclosure with reference to the drawings. Although some embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms. It should be understood that the drawings and embodiments of the present disclosure are only used for exemplary purposes.

[0012] It should be understood that each step described in the embodiments of the method of the present disclosure may be executed in a different order and / or in parallel. In addition, the embodiments of the method may include additional steps and / or omit the execution of the steps shown.

[0013] As used herein, the terms "including" and its variants are open-ended, that is, "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" represents "at least one embodiment", the term "another embodiment" represents "at least one another embodiment", and the term "some embodiments" represents "at least some embodiments". Related definitions of other terms are given in the following description.

[0014] It should be noted that concepts such as "first", "second", etc. mentioned in the present disclosure are only used to distinguish different devices, modules or units.

[0015] It should be noted that the modifiers "one" and "a plurality of" mentioned in the present disclosure are general, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, they should be understood as "one or more".

[0016] In each of the following embodiments, selective features and examples are provided simultaneously in each embodiment. Each feature described in the embodiment may be combined to form a plurality of alternative solutions, and each numbered embodiment should not be regarded as only one technical solution.

[0017] FIG. 1 is a flowchart of a video playback method according to an embodiment of the present disclosure. This method may be executed by a video playback device, where this device may be implemented by at least one of software and hardware, and generally may be integrated into an electronic device. The electronic device may be a server or other devices such as a mobile phone or a tablet computer. As shown in FIG. 1, this method includes the following.

[0018] Step 101, detecting that a preset application for playing a video stream is switched from a non-background running state to a background running state during the playback of a first video.

[0019] In the embodiments of the present disclosure, the type of the preset application may be a short video application or other types of applications. The preset application can provide a video stream push function, and the source of the video stream, etc., may be set according to actual needs by the preset application. For example, it may include a video uploaded by a user who is a video creator to the service end, and may further include a video produced on the platform side corresponding to the preset application. The preset application is generally installed on a terminal used by the user as a client, and may provide a video stream push service to the user in cooperation with the corresponding service end.

[0020] Exemplarily, when running on the terminal to which the application belongs, the running state generally includes a non-background running state and a background running state. When the preset application is in the non-background running state, generally it occupies the entire terminal screen and can display a video playback page. At this time, the user can view the screen content of the video by looking at the screen, and can also input related playback control operations or other operations based on the video playback page. When the preset application is in the background running state, the terminal is in a state such as screen off or screen locked, or there may be other applications running in the non-background. That is, at this time, what is displayed on the screen is the interface of other applications, or the terminal is displaying the desktop. At this time, the preset application may run in the form of a background service or a background process. Generally, it cannot display the normal page of the preset application on the screen and may not display the related content of the preset application. It is also possible to display the preset application in the form of a movable control that occupies a small amount of screen display space such as a floating window or a floating control bar.

[0021] Exemplarily, the preset application may play a first video in the non-background running state, and at this time, the screen content of the video can be viewed from the screen. During the playback of the first video, the user may have a need to use other applications, and the preset application may be switched to run in the background.

[0022] In an embodiment of the present disclosure, for a detection method of detecting whether a preset application is switched from a non-background running state to a background running state, exemplarily, the running state information of the preset application is obtained from the system information of the terminal to which the application belongs, and further, it may be determined whether it is detected that the running state of the preset application is switched from a non-background running state to a background running state. Note that the switch from the non-background running state to the background running state is generally executed in response to a user operation. For example, the user may trigger the switch by triggering a control for switching to the background. In such a case, it may be detected whether this control is triggered, and further, it may be determined whether it is detected that the running state of the preset application is switched from a non-background running state to a background running state.

[0023] Step 102, continue to play the first video in the background by the preset application, and determine a second video from a second video candidate set based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model. Here, the preset background video recommendation model is used to determine the background playback video to be recommended, the second video candidate set is used to select a video from it in the background playback mode, the second video candidate set is different from the first video candidate set, and the first video candidate set is used to select a video from it in the non-background playback mode.

[0024] In an embodiment of the present disclosure, when switching from a non-background playback state to a background playback state, the user can obtain more complete related information of this video. To ensure the video consumption experience, a video that is not being played in the background may continue to be played in the background. Continuing to play may be understood as determining a time point corresponding to the current playback progress in the first video according to the time when switching from the non-background running state to the background running state, and continuing to play the audio corresponding to the first video from the determined time point.

[0025] In an embodiment of the present disclosure, after continuing to play the first video, a video suitable for background playback is determined as the next video of the first video in the video stream. Exemplarily, when making video recommendations, the preset application generally selects from a set of video candidates (which may be understood as a video candidate pool). In the related art, some applications select and play videos from the set of video candidates when running in the non-background, and after switching from the non-background running state to the background running state, stop playing and do not continue to recommend videos. On the other hand, in an embodiment of the present disclosure, to ensure that the preset application can realize continuous video recommendations in the background, a background playback mode is set, and a second set of video candidates is configured for the background playback mode to select a video from it and perform background playback when running in the background.

[0026] In the embodiments of the present disclosure, for non-background video playback and background video playback, corresponding video recommendation models, namely a preset non-background video recommendation model and a preset background video recommendation model, may be set respectively. The preset non-background video recommendation model corresponds to a first set of video candidates, and the first set of video candidates is used to select a video therefrom during non-background playback mode. Generally, the preset non-background video recommendation model is used to determine the non-background playback video to be recommended from the first set of video candidates. The preset background video recommendation model corresponds to a second set of video candidates, and the second set of video candidates is used to select a video therefrom during background playback mode. Generally, the preset background video recommendation model is used to determine the background playback video to be recommended from the second set of video candidates. The determination methods of the preset non-background video recommendation model and the first set of video candidates may be the same as or different from the implementation methods in the related art. The preset background video recommendation model may be a neural network model, and the actual model structure, parameters, training process, etc. may be set according to the actual situation.

[0027] Generally speaking, a video includes image content and audio content. When a user obtains screen information and audio information simultaneously, the user can better understand the whole video. However, when the preset application is in the background running state, the user cannot obtain the screen information, that is, the image content is missing. For a video rich in image content, it is difficult to effectively understand the video content only with the audio content. Generally, there are certain differences among different videos in the video stream. In the embodiments of the present disclosure, based on the characteristics of background playback, the videos in the second set of video candidates may be set to include videos more suitable for listening.

[0028] Exemplarily, the videos in the second video candidate set satisfy preset conditions, which may be set based on whether the videos are suitable for the user to listen to, that is, the preset conditions may include that the videos are suitable for the user to listen to. Optionally, the preset conditions include at least one of matching a preset video type, the ratio of the total time length of the video including voice content to the total time length of the video being greater than a preset ratio threshold, and the total time length of the video being greater than a preset time length threshold.

[0029] Exemplarily, the classification method and classification granularity of video types, etc., may be set according to the actual situation, and may include, for example, comedy, talk shows, skits, singing, dancing, scenic tourism, and sharing of daily life, etc. Here, comedy, talk shows, and singing, etc., may include many vocal performances, and the lack of screen information may have a relatively small impact on the overall understanding of the video, so they may be set as preset video types. Dancing may include many body performances, and scenic tourism may include many scenic spots, etc. Since the lack of screen information may have a relatively large impact on the overall understanding of the video, similar types may be excluded from the preset video types. Skits and sharing of daily life may include a lot of screen information and voice information at the same time, and it may be determined whether to be set as a preset video type according to actual needs, or other elements may be combined to determine whether to be included in the second video candidate set.

[0030] Exemplarily, when there is relatively a large amount of voice content in a video, the voice information is relatively rich. Even if the screen information is missing, for the user, there is a certain listening value. Therefore, a video in which the ratio of the time length of the video containing voice content to the total time length of the video is greater than the preset ratio threshold may be included in the second video candidate set. Here, the time length containing voice content may refer to the sum of the time lengths in which one or more constituent parts (which may be understood as segments) of the voice content are included in the video. A voice recognition technology may be adopted to identify voice components, and further determine the time length containing voice content. For example, a video may be identified by using a Voice Activity Detection (VAD) model. The function of the VAD model is to identify voice segments and non-voice segments from a voice signal, and identify one or more voice segments from the audio data corresponding to the video by using the VAD model, and calculate the sum of the time lengths of the identified voice segments, which may include obtaining the time length containing voice content. The preset ratio threshold may be set to, for example, 60% according to actual needs.

[0031] Exemplarily, since the expression ability of the screen information is stronger, when it is expected that the total time length of the video is relatively short during the creation or production of the video, in order to attract other users to view this video, generally, the information expected to be shown to the user is embodied on the screen as much as possible. Therefore, the screen content of such a video is also relatively rich. Conversely, when it is expected that the total time length of the video is relatively long, the screen content may not need to be rich. In the embodiments of the present disclosure, based on the above idea, preset conditions may be set according to the total time length of the video. For example, when the total time length of the video is greater than the preset time length threshold, the preset conditions are satisfied. Here, the preset time length threshold may be determined according to the range of the video length allowed to be played in the preset application, and may be, for example, the median value of this range.

[0032] Optionally, at least two of the above conditions may be combined, or other conditions may be combined to set preset conditions in order to more accurately select videos suitable for listening and add them to the second set of video candidates.

[0033] In an embodiment of the present disclosure, the current user may be understood as a user using a preset application, and the historical operation information of the current user with respect to the preset application may be understood as relevant information of various operations performed by the current user on the preset application within a preset time period before the current time. Here, the preset time period may be set according to actual needs, for example, it may be set to one month retroactively from the current time. The operations performed by the user may include, for example, click, like, comment, favorite, transfer, pause, switch to the next, switch to the previous, and view the video creator's homepage, etc., and the relevant information of the operation may include, for example, the operation frequency, the operation time (specific time, for example, it may include 8 am, relative time, for example, it may further include the time relative to the video playback start point of the switch operation, for example, when the current video is played for 1 minute and then switched to the next, thereby determining the playback time length of the video) and the operation target (for example, which video or what type of video the operation is performed on), etc., and the relevant information of the operation may further include a state in which the preset application enters because the user does not perform an operation, for example, because the user does not switch to the next during the current video playback, the current video is completely played and enters a repeated playback state, or automatically switches to the playback of the next video, etc.

[0034] Exemplarily, the historical operation information may include non-background historical operation information and background historical operation information. The non-background historical operation information may be understood as related information of various operations performed in the non-background, and the background historical operation information may be understood as related information of various operations performed in the background. Generally, background operations are relatively few and may include, for example, pausing, switching to the next, switching to the previous, and favorites, etc., and the related information may include, for example, whether to completely play a video and the video playback duration, etc.

[0035] The historical operation information of the current user with respect to the preset application can reflect some preferences of the current user for different videos. In the embodiments of the present disclosure, the historical operation information may be applied to video recommendation when obtaining the user's consent, thereby more personalized determining the video to be played in the background for the current user based on the historical operation information and the preset background video recommendation model.

[0036] Exemplarily, the historical operation information and the feature information of each candidate video in the second video candidate set (for example, related information such as video tags, video creators, video duration, video popularity, video image information, and video audio information may also be included) may be input as input data into the preset background video recommendation model, and the second video may be determined based on the output result of the preset background video recommendation model.

[0037] Step 103, when detecting a preset video switching event, the preset application plays the second video in the background.

[0038] Exemplarily, in the process of continuously playing the first video in the background, when a preset video switching event is detected, the preset application may play the second video in the background. Exemplarily, video switching in the background playback mode may be realized by playing the audio corresponding to the second video. The preset video switching event may be understood as an event that triggers background video switching, may be automatically triggered by the system, may be triggered by the user spontaneously, or may be set according to actual needs.

[0039] In the video playback method according to the embodiments of the present disclosure, the preset application detects that it is switched from the non-background running state to the background running state during the playback of the first video, and the preset application continues to play the first video in the background. Based on the historical operation information of the current user for the preset application and the preset background video recommendation model, a second video is determined from the second video candidate set. When a preset video switching event is detected, the preset application plays the second video in the background. By adopting the above embodiment, the preset application can distinguish video recommendations when it is in the non-background running state and the background running state respectively. After being switched from the non-background running state to the background running state, the video being played in the background is continuously played, a video recommendation model and a video candidate set different from the non-background running state are adopted, and with reference to the historical operation information of the current user, a video suitable for background playback is determined and played when a preset video switching event is detected, thereby ensuring the continuity of the video information obtained by the user, and performing more accurate personalized video recommendations in the background running state, and improving the performance of the application.

[0040] In some embodiments, after the preset application starts playing the second video in the background, when a preset video switching event is detected, the preset application further includes playing a third video in the background, where the third video is determined from the set of second video candidates based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model. The advantage of such a setting is that it can support the switching and playing among multiple videos determined by the preset background video recommendation model. Here, the historical operation information used as a basis when determining the third video may be the same as or different from the historical operation information used as a basis when determining the second video. When the historical operation information used as a basis when determining the third video is the same as the historical operation information used as a basis when determining the second video, it may be understood that multiple background playback videos to be recommended may be collectively determined based on the historical operation information and the preset background video recommendation model, and it may also be a single video queue. When a preset video switching event is detected, the next video in the video queue is played to ensure the push efficiency. When the historical operation information used as a basis when determining the third video is different from the historical operation information used as a basis when determining the second video, the historical operation information is updated based on the operation information during the playback of the second video, and the third video is determined based on the updated historical operation information and the preset background video recommendation model, which may be understood as making the recommendation result more accurate.

[0041] In some embodiments, detecting the preset video switching event may include detecting that the playback of the current video has been completed. The advantage of such a setting is that after the playback of the current video is completed, the next video can be automatically played. Since the preset application is in the background running state, at this time, it may be difficult for the user to operate the preset application. Automatic playback can reduce user operations while assisting the user to obtain new video content, and can improve video push efficiency and platform resource utilization rate.

[0042] In some embodiments, detecting the preset video switching event may include receiving a video switching command input by the current user. The advantage of such a setting is that when the current user has no interest in the currently playing video, it can quickly switch to the playback of the next video and improve the interaction efficiency. Optionally, when a floating control bar corresponding to the preset application is displayed on the screen, a video switching command input by the current user may be received based on the floating control bar. For example, the floating control bar may include a "Next" button, and the user can input a video switching command by clicking this button. Optionally, the video switching command may not require the display of the floating control bar and may be input by a voice method that saves screen display space, and it may also be applicable when there is a switching need when the user is at a certain distance from the terminal, for example, when putting down the mobile phone to do other things.

[0043] In some embodiments, the historical operation information includes non-background historical operation information and background historical operation information. Here, determining a second video from a second set of video candidates based on the historical operation information of the current user for the preset application and the preset background video recommendation model includes inputting the non-background historical operation information into the preset non-background video recommendation model and obtaining an intermediate output result in the preset non-background video recommendation model, which is used to determine the non-background playback video to be recommended by the preset non-background video recommendation model, and the intermediate output result includes the user characteristic information of the current user, and determining a second video from the second set of video candidates based on the user characteristic information of the current user, the background historical operation information, and the preset background video recommendation model. Here, the intermediate output result may be understood as the output result of the intermediate layer of the preset non-background video recommendation model. Exemplarily, the user characteristic information may be understood as user image information, may exist in the form of a feature vector, etc., and can embody the preference information of the user in some aspects, such as preferring to listen to a talk show of an actor. Such preference information may be understood as the personalized information of the current user and generally does not change due to changes in the running state of the preset application in a short time. The intermediate output result may include other information that is advantageous for accurately determining the video to be recommended by the preset background video recommendation model.The advantages of such a setting are as follows: The non-background history operation information is richer than the background history operation information, and the user characteristic information can be determined quickly and accurately. On the other hand, the background history operation information is more suitable for the background playback scenario. By combining the two as the input of the preset background video recommendation model, the background playback video to be recommended that suits the current user can be determined more accurately. In addition, it is beneficial to simplify the preset background video recommendation model and avoid excessive occupation of background resources.

[0044] In some embodiments, determining the second video from the second video candidate set based on the history operation information of the current user for the preset application and the preset background video recommendation model includes: determining a second video candidate subset from the second video candidate set corresponding to the preset background video recommendation model based on the non-background history operation information and the preset non-background video recommendation model; and determining the second video from the second video candidate subset based on the background history operation information and the preset background video recommendation model. The advantage of such a setting is that the non-background history operation information and the preset non-background video recommendation model are used to initially screen the videos in the second video candidate set to obtain the second video candidate subset, and then the background history operation information and the preset background video recommendation model are used to determine the second video from the second video candidate subset, so as to make full use of the user's history operation information in the non-background and background, combine the features of the two video recommendation models, and quickly and accurately determine the video suitable for playback in the background that conforms to the user's preferences.

[0045] In some embodiments, the training data used in the training phase by the preset background video recommendation model includes at least one of a user sample corresponding to the preset non-background video recommendation model and user feature information. The advantage of such a setting is that since the non-background recommended videos are more abundant and the user operation information is also more abundant, at least one of the user sample corresponding to the preset non-background video recommendation model and the user feature information can more completely and accurately reflect the personalized characteristics of the user. By using at least one of the user sample corresponding to the preset non-background video recommendation model and the user feature information as training data, the preset background video recommendation model obtained through training can more accurately identify user features and further determine a background recommended video that matches the user features.

[0046] In some embodiments, determining the second video from the second set of video candidates based on the historical operation information of the current user for the preset application and the preset background video recommendation model includes determining the second video from the second set of video candidates based on the non-background historical operation information and the preset non-background video recommendation model when it is determined that the data amount of the background historical operation information is smaller than the preset data amount threshold corresponding to the preset background video recommendation model. The advantage of such a setting is that when using the preset background video recommendation model for video recommendation, generally, it is necessary to accumulate user operation information. For users who rarely perform background playback or at a stage where the background video playback function has just been used, the background historical operation information is often relatively small, which affects the accuracy of the recommendation result. At this time, the second video may be determined from the second set of video candidates by first using the non-background historical operation information and the preset non-background video recommendation model. Even in the cold start stage, it is possible to provide a personalized background video recommendation function for the user.

[0047] In some embodiments, the second set of video candidates includes a first type of video and a second type of video. The amount of image information in the first type of video is less than or equal to a first preset information amount threshold. The amount of image information in the second type of video is greater than a second preset information amount threshold, and the second preset information amount threshold is greater than or equal to the first preset information amount threshold. Here, for the first type of video, the input data when using the preset background video recommendation model includes the audio information in the first type of video. For the second type of video, the input data when using the preset background video recommendation model includes the image information and audio information in the first type of video. The advantage of such a setting is that videos can be classified in terms of the richness of screen information, different inputs can be adopted for different types of videos, and the recommendation results of the model can be made more accurate.

[0048] Exemplarily, during video creation, if more emphasis is placed on audio creation, for example, there may be cases where one's perspective on a personal hot event is described or a story is told. At this time, the arranged images are one or more small still images, with relatively insufficient screen information, and the image content may not be very helpful for understanding the audio content. Since such videos are closer to audio, this type of video may be classified as the first type of video. The amount of image information in a video may be determined based on the sum of the amount of image information in each video frame (excluding overlapping video frames), and the amount of image information may be calculated by a method such as information entropy.

[0049] When performing video recommendation for the first type of video with a relatively small amount of image information using a preset background video recommendation model, the image information may be ignored. However, when performing video recommendation for the second type of video with a relatively large amount of image information using the preset background video recommendation model, the image information and audio information may be considered simultaneously. In this way, more accurate background recommendation can be performed for different types of videos.

[0050] Optionally, as one implementation method, different background video recommendation models, for example, the first preset background video recommendation model and the second background video recommendation model, may be set for the first type of video and the second type of video respectively. The two models are trained and used separately. When background video recommendation needs to be performed, a first number of first videos to be recommended are determined based on the historical operation information and the first preset background video recommendation model, and a second number of second videos to be recommended are determined based on the historical operation information and the second preset background video recommendation model. The first videos to be recommended and the second videos to be recommended are sorted in a mixed manner to obtain a video queue of videos to be recommended, and then the next video that needs to be played in the background is sequentially determined from the video queue of videos to be recommended.

[0051] Optionally, for the first type of video, it may not be suitable for non-background playback. When the first type of video is not included in non-background video recommendation, after entering the background operation state, there may be no available non-background operation information. At this time, corresponding cold start processing may be performed on the first type of video. For example, the first type of video with a relatively high current popularity ranking may be determined as the first video to be recommended.

[0052] Figure 2 is a flowchart of another video playback method according to an embodiment of the present disclosure. The embodiment of the present disclosure is changed based on a plurality of alternative solutions in the above embodiment. This method includes the following steps.

[0053] Step 201, detect that the preset application is switched from the non-background running state to the background running state during the playback of the first video.

[0054] Exemplarily, the preset application is a short video application. When the user is browsing a short video stream using the short video application and there is a need to operate other applications, the preset application may be switched to run in the background. For example, when currently playing short video A, the user may slide from the bottom of the screen to enter the desktop of the terminal. At this time, a floating control bar may be displayed on the desktop interface. The floating control bar may include sub-controls such as a pause button and a next button. Then, the user may start other applications.

[0055] Step 202, continue to play the first video in the background by the preset application.

[0056] After switching the short video application to run in the background, short video A may continue to be played in the background.

[0057] Exemplarily, assuming that the total duration of short video A is 3 minutes, when the short video A is played for 1 minute, the user may switch the short video application to run in the background, record the time point of 1 minute at this time, and continue to play the audio corresponding to short video A after 1 minute.

[0058] Step 203: Input the non-background history operation information of the current user for the preset application into the preset non-background video recommendation model, and obtain the intermediate output result in the preset non-background video recommendation model. Here, the intermediate output result includes the user feature information of the current user.

[0059] Exemplarily, the non-background history operation information may include the operation information when the current user browsed the shot video stream using the shot video application in the non-background within the past one month. Input the non-background history operation information into the preset non-background video recommendation model. It should be noted that when using the preset non-background video recommendation model, other data may also need to be input. Then, obtain the user feature information of the current user from the output result of the intermediate layer of the preset non-background video recommendation model.

[0060] Step 204: Determine the second video from the second video candidate set based on the user feature information, the background history operation information, and the preset background video recommendation model.

[0061] Exemplarily, the user feature information of the current user, the background history operation information, and the feature information of each candidate video in the second video candidate set may be input into the preset background video recommendation model as input data, and the second video may be determined based on the output result of the preset background video recommendation model.

[0062] Here, the videos in the second video candidate set meet the preset conditions. The preset conditions include at least one of the following: matching the preset video type, the ratio of the total time length of the video containing voice content to the total time length of the video being greater than the preset ratio threshold, and the total time length of the video being greater than the preset time length threshold.

[0063] When it is detected in step 205 that the playback of the first video has been completed, start playing the second video in the background.

[0064] It should be noted that if a video switching command input by the current user is received before the playback of the first video is completed, the playback of the second video may be started in the background.

[0065] When it is detected in step 206 that a video switching command input by the current user is received, start playing the third video in the background, where the third video is determined from a set of second video candidates based on the historical operation information of the current user for the preset application and the preset background video recommendation model.

[0066] Exemplarily, when the user is using another non-background application and is not interested in the audio content of the currently playing second video, the user may complete the input of the video switching command by clicking the next button on the floating control bar, thereby realizing the switching to the background playback of the next video.

[0067] It should be noted that if a video switching command input by the current user is not received when the playback of the second video is completed, the playback of the third video may be automatically started in the background.

[0068] According to the video playback method according to an embodiment of the present disclosure, in the process of a preset application playing a video stream in the foreground, when the user switches the preset application to run in the background, the background playback of the current video in the audio format is continued, and the user's foreground history operation information is combined with the preset foreground video recommendation model to determine user characteristic information. Using the user characteristic information, the background history operation information, and the preset background video recommendation model, a video stream suitable for background playback can be continuously recommended to the user. During background playback, continuous automatic playback can be realized, supporting the user to input a video switching command, providing the user with a new content consumption form, improving the user experience, and enhancing the performance of the application. It is also possible to improve the exposure rate of videos on the platform and improve the utilization rate of platform resources.

[0069] FIG. 3 is a structural block diagram of a video playback device according to an embodiment of the present disclosure. This device may be implemented by software and / or hardware, may generally be integrated into an electronic device, and may perform video playback by executing a video playback method. As shown in FIG. 3, this device includes a state detection module 301 arranged to detect whether a preset application for playing a video stream has been switched from a foreground running state to a background running state during the playback of a first video; After detecting that the preset application is switched from the non-background running state to the background running state during the playback of the first video, the preset application continues to play the first video in the background, and is arranged to determine a second video from a set of second video candidates based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model. The preset background video recommendation model is used to determine the background playback video to be recommended, the set of second video candidates is used to select a video therefrom during the background playback mode, the set of second video candidates is different from the set of first video candidates, and the set of first video candidates is used to select a video therefrom during the non-background playback mode. A video determination module 302, A playback switching module 303 arranged to play the second video in the background by the preset application when a preset video switching event is detected.

[0070] The video playback device according to the embodiments of the present disclosure distinguishes video recommendations in the non-background running state and the background running state of each preset application. After being switched from the non-background running state to the background running state, the video being played in the background is continued to be played, a video recommendation model and a set of video candidates different from the non-background running state are adopted, and with reference to the historical operation information of the current user, a video suitable for background playback is determined and played when a preset video switching event is detected, thereby guaranteeing the continuity of the video information obtained by the user, and performing more accurate personalized video recommendations in the background running state, and improving the performance of the application.

[0071] Optionally, the videos in the second video candidate set satisfy preset conditions, and the preset conditions include at least one of matching a preset video type, the ratio of the time length of the video including voice content to the total time length of the video being greater than a preset ratio threshold, and the total time length of the video being greater than a preset time length threshold.

[0072] Optionally, after the playback switching module further starts playing the second video in the background by the preset application, when a preset video switching event is detected, it is arranged to play a third video in the background by the preset application, where the third video is determined from the second video candidate set based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model.

[0073] Optionally, detecting the preset video switching event includes detecting that the playback of the current video has been completed or receiving a video switching command input by the current user.

[0074] Optionally, the historical operation information includes non-background historical operation information and background historical operation information. After detecting that the preset application is switched from a non-background running state to a background running state during the playback of the first video, the video determination module is arranged to continuously play the first video in the background by the preset application, input the non-background historical operation information into a preset non-background video recommendation model, and obtain an intermediate output result of the preset non-background video recommendation model, which is used to determine the non-background playback video recommended by the preset non-background video recommendation model, and the intermediate output result includes user feature information of the current user. The video determination unit is configured to determine a second video from a second set of video candidates based on the user feature information of the current user, the background historical operation information, and the preset background video recommendation model.

[0075] Optionally, the training data used in the training stage by the preset background video recommendation model includes at least one of a user sample corresponding to the preset non-background video recommendation model and user feature information.

[0076] Optionally, when the video determination unit determines that the data volume of the background historical operation information is smaller than a preset data volume threshold corresponding to the preset background video recommendation model, the video determination unit is arranged to determine a second video from a second set of video candidates based on the non-background historical operation information and the preset non-background video recommendation model.

[0077] Optionally, the second video candidate set includes a first type of video and a second type of video. The amount of image information in the first type of video is less than or equal to a first preset information amount threshold, and the amount of image information in the second type of video is greater than a second preset information amount threshold, where the second preset information amount threshold is greater than or equal to the first preset information amount threshold. For the first type of video, the input data when using the preset background video recommendation model includes the audio information in the first type of video. For the second type of video, the input data when using the preset background video recommendation model includes the image information and audio information in the first type of video.

[0078] Hereinafter, with reference to FIG. 4, a schematic structural diagram of an electronic device 400 suitable for implementing the embodiments of the present disclosure is shown. The electronic device in the embodiments of the present disclosure may include mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, personal digital assistants (PDAs), tablet computers (Portable Android Devices, PADs), portable multimedia players (PMPs), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The electronic device shown in FIG. 4 is merely an example.

[0079] As shown in FIG. 4, the electronic device 400 may include a processing device (e.g., a central processor, a graphics processor, etc.) 401 that can execute various appropriate operations and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a storage device 408 into a random access memory (RAM) 403. In the RAM 403, various programs and data necessary for the operation of the electronic device 400 are stored. The processing device 401, the ROM 402, and the RAM 403 are connected to each other via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0080] Generally, an input device 406 including, for example, a touch screen, a touch panel, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyro, etc., an output device 407 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc., a storage device 408 including, for example, a magnetic tape, a hard disk, etc., and a communication device 409 may be connected to the I / O interface 405. The communication device 409 may allow the electronic device 400 and other devices to perform wireless or wired communication to exchange data. FIG. 4 shows an electronic device 400 having various devices, but it should be understood that it is not required to implement or include all the devices shown. Alternatively, more or fewer devices may be implemented or included.

[0081] In particular, according to the embodiments of the present disclosure, the processes described with reference to the above-mentioned flowcharts may be implemented as computer software programs. For example, the embodiments of the present disclosure include a computer program product, which includes a computer program mounted on a non-transitory computer-readable medium, and this computer program includes program code for executing the method shown in the flowchart. In such an embodiment, this computer program may be downloaded and installed from a network via the communication device 409, or installed from the storage device 408, or installed from the ROM 402. When this computer program is executed by the processing device 401, the above functions in the method of the embodiments of the present disclosure are executed.

[0082] It should be noted that the above computer-readable medium according to the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. The computer-readable storage medium may include an electrical connection having one or more conductors, a portable computer magnetic disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical memory device, a magnetic memory device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium may be any tangible medium that includes or stores a program, and this program may be used by or in combination with an instruction execution system, apparatus, or device. On the other hand, in the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier, in which computer-readable program code is carried. The data signal propagated in this way may adopt various forms including electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may be any computer-readable medium other than the computer-readable storage medium, and this computer-readable signal medium may transmit, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code included on the computer-readable medium may be transmitted through any suitable medium including electric wires, optical cables, radio frequency (RF), etc., or any suitable combination of the above.

[0083] The computer-readable medium may be included in the electronic device or may exist independently without being assembled into the electronic device.

[0084] One or more programs are installed on the computer-readable medium. When the one or more programs are executed by the electronic device, the electronic device detects that a preset application for playing a video stream is switched from a non-background running state to a background running state during the playback of a first video, and continuously plays the first video in the background by the preset application. Based on the historical operation information of the current user for the preset application and a preset background video recommendation model, a second video is determined from a second set of video candidates. The preset background video recommendation model is used to determine the background playback video to be recommended, and the second set of video candidates is used to select a video from it in the background playback mode. The second set of video candidates is different from the first set of video candidates, and the first set of video candidates is used to select a video from it in the non-background playback mode. When a preset video switching event is detected, the preset application is enabled to play the second video in the background.

[0085] Computer program code for performing the operations of the present disclosure may be created by one or more programming languages or combinations thereof, and the programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and further include conventional procedural programming languages such as the "C" language or similar programming languages. The program code may be executed entirely on a user computer, partially executed on a user computer, executed as one independent software package, partially executed on a user computer, partially executed on a remote computer, or executed entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user computer via any type of network including a Local Area Network (LAN) or a Wide Area Network (WAN), or may be connected to an external computer (for example, connected via the Internet through an Internet service provider).

[0086] The flowcharts and block diagrams in the drawings illustrate the possible system architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, a program segment, or a portion of code that includes one or more executable instructions for implementing a specified logical function. It should also be noted that in some realizations as alternatives, the functions assigned to the blocks may occur in an order different from that shown in the drawings. For example, two blocks represented consecutively may actually be executed substantially in parallel, or they may be executed in the reverse order, depending on the functions involved. Further, it should be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0087] The modules described in connection with the embodiments of the present disclosure may be implemented in software or in hardware. Here, when the name of a module is one, it does not limit the module itself. For example, the playback switching module may be described as "a module that plays the second video in the background by the preset application when detecting a preset video switching event."

[0088] The functions described above in this specification may be executed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that may be used include Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Parts (ASSPs), System on Chips (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.

[0089] In the context of this disclosure, a machine-readable medium may be a tangible medium that includes, or stores, a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium may be either a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. A machine-readable storage medium may include, among other things, electrical connections based on one or more lines, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fiber, compact disc read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0090] According to one or more embodiments of the present disclosure, Example 1 provides a video playback method, detecting that a preset application for playing a video stream is switched from a non-background running state to a background running state during the playback of a first video; Continuously play the first video in the background by the preset application, and determine a second video from a set of second video candidates based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model, where the preset background video recommendation model is used to determine the background playback video to be recommended, the set of second video candidates is used to select a video therefrom during the background playback mode, the set of second video candidates is different from the set of first video candidates, and the set of first video candidates is used to select a video therefrom during the non-background playback mode; When a preset video switching event is detected, include playing the second video in the background by the preset application.

[0091] According to one or more embodiments of the present disclosure, in Example 2, based on the method described in Example 1, the videos in the set of second video candidates satisfy preset conditions, and the preset conditions include at least one of matching a preset video type, the ratio of the total time length of the video including voice content to the total time length of the video being greater than a preset ratio threshold, and the total time length of the video being greater than a preset time length threshold.

[0092] According to one or more embodiments of the present disclosure, in Example 3, based on the method described in Example 1, after starting to play the second video in the background by the preset application, When a preset video switching event is detected, further include playing a third video in the background by the preset application, where the third video is determined from the set of second video candidates based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model.

[0093] According to one or more embodiments of the present disclosure, in Example 4, based on the method described in Example 1, detecting the preset video switching event means that it is detected that the playback of the current video has been completed, or receiving a video switching command input by the current user is included.

[0094] According to one or more embodiments of the present disclosure, in Example 5, based on the method described in any one of Examples 1 to 4, the historical operation information includes non-background historical operation information and background historical operation information, wherein, based on the historical operation information of the current user for the preset application and the preset background video recommendation model, determining a second video from a second set of video candidates inputting the non-background historical operation information into a preset non-background video recommendation model to obtain an intermediate output result of the preset non-background video recommendation model, which is used to determine a non-background playback video to be recommended by the preset non-background video recommendation model, and the intermediate output result includes user feature information of the current user, and determining a second video from a second set of video candidates based on the user feature information of the current user, the background historical operation information, and the preset background video recommendation model.

[0095] According to one or more embodiments of the present disclosure, in Example 6, based on the method described in Example 5, the training data used in the training stage by the preset background video recommendation model includes at least one of a user sample corresponding to the preset non-background video recommendation model and user feature information.

[0096] According to one or more embodiments of the present disclosure, Example 7 is based on the method described in Example 5, and determining a second video from a second set of video candidates based on the historical operation information of the current user for the preset application and the preset background video recommendation model, when it is determined that the data amount of the background historical operation information is smaller than a preset data amount threshold corresponding to the preset background video recommendation model, determining a second video from the second set of video candidates based on the non-background historical operation information and the preset non-background video recommendation model.

[0097] According to one or more embodiments of the present disclosure, Example 8 is based on the method described in Example 1. The second set of video candidates includes a first type of video and a second type of video. The amount of image information in the first type of video is equal to or less than a first preset information amount threshold, and the amount of image information in the second type of video is greater than a second preset information amount threshold, and the second preset information amount threshold is equal to or greater than the first preset information amount threshold. Here, for the first type of video, the input data when using the preset background video recommendation model includes the audio information in the first type of video. For the second type of video, the input data when using the preset background video recommendation model includes the image information and audio information in the first type of video.

[0098] According to one or more embodiments of the present disclosure, Example 9 provides a video playback device. A state detection module arranged to detect whether a preset application for playing a video stream has been switched from a non-background running state to a background running state during the playback of a first video. A video determination module that, after detecting that a preset application is switched from a non-background running state to a background running state during playback of a first video, continues to play the first video in the background by the preset application, and determines a second video from a set of second video candidates based on the historical operation information of the current user with respect to the preset application and a preset background video recommendation model. The preset background video recommendation model is used to determine the background playback video to be recommended, the set of second video candidates is used to select a video therefrom during the background playback mode, the set of second video candidates is different from the set of first video candidates, and the set of first video candidates is used to select a video therefrom during the non-background playback mode. And a playback switching module arranged to play the second video in the background by the preset application when a preset video switching event is detected.

Claims

1. A video playback method, comprising: detecting that a preset application for playing a video stream is switched from a non-background running state to a background running state during playback of a first video; continuing to play the first video in the background by the preset application, and determining a second video from a second set of video candidates based on historical operation information of the current user with respect to the preset application and a preset background video recommendation model, wherein the preset background video recommendation model is used to determine a background playback video to be recommended, the second set of video candidates is used to select a video therefrom in a background playback mode, the second set of video candidates is different from a first set of video candidates, and the first set of video candidates is used to select a video therefrom in a non-background playback mode; when detecting a preset video switching event, playing the second video in the background by the preset application; the second set of video candidates includes a first type of video and a second type of video, an amount of image information in the first type of video is equal to or less than a first preset information amount threshold, an amount of image information in the second type of video is greater than a second preset information amount threshold, and the second preset information amount threshold is equal to or greater than the first preset information amount threshold; here, for the first type of video, input data when using the preset background video recommendation model includes audio information in the first type of video; for the second type of video, input data when using the preset background video recommendation model includes image information and audio information in the first type of video. A video playback method.

2. The videos in the second video candidate set satisfy preset conditions, where the preset conditions include at least one of matching the preset video type, the ratio of the time length of the video including voice content to the total time length of the videos being greater than the preset ratio threshold, and the total time length of the video being greater than the preset time length threshold. The method according to claim 1.

3. After starting the playback of the second video in the background by the preset application, when a preset video switching event is detected, further including playing a third video in the background by the preset application, where the third video is determined from the second video candidate set based on the historical operation information of the current user with respect to the preset application and the preset background video recommendation model. The method according to claim 1.

4. Detecting the preset video switching event includes detecting that the playback of the current video has been completed, or receiving a video switching command input by the current user. The method according to claim 1.

5. The historical operation information includes non-background historical operation information and background historical operation information, and determining a second video from the second video candidate set based on the historical operation information of the current user with respect to the preset application and the preset background video recommendation model includes inputting the non-background historical operation information into a preset non-background video recommendation model to obtain an intermediate output result of the preset non-background video recommendation model, where the intermediate output result is used to determine the non-background playback video to be recommended by the preset non-background video recommendation model, and the intermediate output result includes the user feature information of the current user, and determining a second video from the second video candidate set based on the user feature information of the current user, the background historical operation information, and the preset background video recommendation model. The method according to claim 1.

6. The training data used in the training phase by the preset background video recommendation model includes at least one of the user sample corresponding to the preset non-background video recommendation model and the user feature information, according to the method of claim 5.

7. Determining a second video from a second set of video candidates based on the historical operation information of the current user for the preset application and the preset background video recommendation model includes: When it is determined that the data volume of the background historical operation information is smaller than the preset data volume threshold corresponding to the preset background video recommendation model, determining a second video from the second set of video candidates based on the non-background historical operation information and the preset non-background video recommendation model, according to the method of claim 5.

8. A video playback device, A state detection module arranged to detect whether a preset application for playing a video stream has been switched from a non-background running state to a background running state during the playback of a first video; A video determination module arranged to continue playing the first video in the background by the preset application after detecting that the preset application has been switched from a non-background running state to a background running state during the playback of the first video, and to determine a second video from a second set of video candidates based on the historical operation information of the current user for the preset application and the preset background video recommendation model, wherein the preset background video recommendation model is used to determine the background playback video to be recommended, the second set of video candidates is used to select a video therefrom during the background playback mode, the second set of video candidates is different from the first set of video candidates, and the first set of video candidates is used to select a video therefrom during the non-background playback mode. A playback switching module arranged to play the second video in the background by the preset application when a preset video switching event is detected. The set of second video candidates includes a first type of video and a second type of video. The amount of image information in the first type of video is less than or equal to a first preset information amount threshold. The amount of image information in the second type of video is greater than a second preset information amount threshold, and the second preset information amount threshold is greater than or equal to the first preset information amount threshold. Here, for the first type of video, the input data when using the preset background video recommendation model includes the audio information in the first type of video. For the second type of video, the input data when using the preset background video recommendation model includes the image information and audio information in the first type of video, a video playback device.

9. An electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the method according to any one of claims 1 to 7 is realized.

10. A computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is realized.

Images