CN114095742A

CN114095742A - Video recommendation method and device, computer equipment and storage medium

Info

Publication number: CN114095742A
Application number: CN202111319368.0A
Authority: CN
Inventors: 莫筱羽; 陈泽宇
Original assignee: Guangzhou Boguan Information Technology Co Ltd
Current assignee: Guangzhou Boguan Information Technology Co Ltd
Priority date: 2021-11-09
Filing date: 2021-11-09
Publication date: 2022-02-25

Abstract

The embodiment of the application discloses a video recommendation method and device, computer equipment and a storage medium. According to the scheme, the content in the game live video stream is identified in multiple dimensions, the identification result is obtained through data including game scenes, game processes and the like, then the target video similar to the content of the current live video is selected from the current game live video according to the identification result, then video recommendation is carried out on the user of the live client according to the target video, the user can watch related videos rapidly, and therefore video recommendation efficiency can be improved.

Description

Video recommendation method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a video recommendation method and apparatus, a computer device, and a storage medium.

Background

With the continuous development of network technologies, emerging live broadcast platforms are more and more popular with users. In order to meet the video watching requirement of the user, the live broadcast platform can recommend videos for the user.

In the related art, most methods for recommending videos to users recommend a series of related whole videos to users based on the types of videos watched by the users, the preference degree and the watching duration of each video type, and the like, but if the videos desired to be watched by the users are partial segments in the whole videos, when the playing duration of the whole videos is long, the users need to spend more time searching for the partial segments, and thus the video recommendation efficiency is low.

Disclosure of Invention

The embodiment of the application provides a video recommendation method and device, a computer device and a storage medium, which can improve video recommendation efficiency.

The embodiment of the application provides a video recommendation method, which comprises the following steps:

acquiring a current game picture of a target game, and live broadcasting the target game through a live broadcasting client;

determining at least one current game element according to the content in the current game picture;

determining a target video from historical live videos of the target game through a current game element;

and displaying the target video on a live interface of the live client.

Correspondingly, the embodiment of the present application further provides a video recommendation device, including:

the first acquisition unit is used for acquiring a current game picture of a target game, and the target game is live broadcast through a live broadcast client;

the first determining unit is used for determining at least one current game element according to the content in the current game picture;

a second determining unit, configured to determine a target video from historical live videos of the target game through the current game element;

and the display unit is used for displaying the target video on a live interface of the live client.

In some embodiments, the second determination unit comprises:

a first determining subunit, configured to determine a scene type of the current game scene in a game map of the target game;

a second determining subunit, configured to determine, from the current virtual game roles, a target virtual game role operated by an anchor user, where the anchor user is a user who performs live broadcast on a target game through the live broadcast client;

and the third determining subunit is used for determining a target video from the historical live video according to the scene type and the target virtual game role.

In some embodiments, the third determining subunit is specifically configured to:

determining a target historical live video matched with the scene type and the target virtual game role from the historical live video;

determining the target video based on the target historical live video.

acquiring at least one marked video clip in the target historical live video;

determining the target video from the marked video segments.

In some embodiments, the third determining subunit is further specifically configured to:

acquiring a historical game picture when game battle performance information of the anchor user in the target historical live video accords with preset battle performance information;

and generating the marked video clip based on the historical game elements in the historical game picture and the historical game progress.

In some embodiments, the second determination unit comprises:

a fourth determining subunit, configured to determine, from the historical live video, a live video corresponding to the anchor user to obtain a candidate live video;

a fifth determining subunit, configured to determine the target video from the candidate live videos through the current game element.

In some embodiments, the second determination unit comprises:

a sixth determining subunit, configured to determine, from the historical live video, a live video corresponding to the target virtual game role, to obtain a candidate live video;

a seventh determining subunit, configured to determine the target video from the candidate live videos through the current game element.

In some embodiments, the second determination unit comprises:

an eighth determining subunit, configured to determine a target position of the current game scene in a game map of the target game;

a ninth determining subunit, configured to determine, from the current virtual game characters, a target virtual game character operated by an anchor user;

and the tenth determining subunit is configured to determine a target video from the historical live video according to the target position and the target virtual game character.

In some embodiments, the apparatus further comprises:

a third determining unit, configured to determine current game information of the target game according to content in the current game screen;

and the fourth determining unit is used for determining a target video from the historical live video of the target game through the current game element and the current game information.

In some embodiments, the third determination unit comprises:

the extracting subunit is used for extracting game battle performance information and game match duration information from the content of the current game picture;

and the eleventh determining subunit is used for determining the game progress based on the game battle performance information and the game play duration information to obtain the current game information.

In some embodiments, the first determination unit comprises:

the identification subunit is used for carrying out scene identification on the current game picture and determining the current game scene in the current game picture;

a twelfth determining subunit, configured to perform role identification on the current game screen, and determine a current virtual game role in the current game screen;

a thirteenth determining subunit, configured to obtain a current game element based on the current game scene and the current virtual game character, and obtain the current game element.

In some embodiments, the apparatus further comprises:

the second acquisition unit is used for acquiring the voice information of the live broadcast client;

the recognition unit is used for recognizing the voice information and determining a keyword of the voice information;

and the execution unit is used for executing the step of determining at least one current game element according to the content in the current game picture if the keywords accord with the keywords recommended by the video.

In some embodiments, the apparatus further comprises:

the third acquisition unit is used for acquiring historical watching behavior information of the user if the keyword does not accord with the keyword recommended by the video;

a fifth determining unit, configured to determine the target video from the historical live video based on the historical viewing behavior information.

In some embodiments, the first determination unit comprises:

a fourteenth determining subunit, configured to determine the current game element according to the keyword and the content in the current game screen.

In some embodiments, the display unit includes:

and the display subunit is used for sending the target video to the live client so that the live client can display the target video on a game live interface of the current game picture.

Accordingly, embodiments of the present application further provide a computer device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the video recommendation method provided in any of the embodiments of the present application.

Correspondingly, the embodiment of the application also provides a storage medium, wherein the storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor to execute the video recommendation method.

According to the method and the device, the identification of multiple dimensions is carried out on the content in the game live video stream, the identification result is obtained through the data including the game scene, the game process and the like, the target video similar to the content of the current live video is selected from the current game live video according to the identification result, then video recommendation is carried out on the user of the live client according to the target video, the user can watch related videos quickly, and therefore video recommendation efficiency can be improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a schematic flowchart of a video recommendation method according to an embodiment of the present application.

Fig. 2 is a schematic flowchart of another video recommendation method according to an embodiment of the present application.

Fig. 3 is a block diagram of a video recommendation apparatus according to an embodiment of the present application.

Fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present application.

Detailed Description

The technical solutions in the embodiments of the present application will be described clearly and completely with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only a part of the embodiments of the present application, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The embodiment of the application provides a video recommendation method, a video recommendation device, a storage medium and computer equipment. Specifically, the video recommendation method according to the embodiment of the present application may be executed by a computer device, where the computer device may be a server or other devices. The server may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as cloud service, a cloud database, cloud computing, a cloud function, cloud storage, network service, cloud communication, middleware service, domain name service, security service, CDN, and a big data and artificial intelligence platform.

For example, the computer device may be a server, and the server may obtain a current game screen of a target game live through a live client; determining at least one current game element according to the content in the current game picture, and determining current game information of the target game according to the content in the current game picture; determining a target video from historical live videos of the target game through current game elements and current game information; and performing video recommendation on the user of the live broadcast client based on the target video.

Based on the above problems, embodiments of the present application provide a video recommendation method, an apparatus, a computer device, and a storage medium, which can improve video recommendation efficiency.

The following are detailed below. It should be noted that the following description of the embodiments is not intended to limit the preferred order of the embodiments.

The embodiment of the present application provides a video recommendation method, which may be executed by a terminal or a server, and is described by taking an example that the video recommendation method is executed by the server.

Referring to fig. 1, fig. 1 is a schematic flowchart illustrating a video recommendation method according to an embodiment of the present disclosure. The specific flow of the video recommendation method can be as follows:

101. and acquiring a current game picture of the target game.

The target game is live broadcast through the live broadcast client.

In the embodiment of the application, the live client can refer to application software with a live function, the live client can be installed on a terminal, and a main broadcast user needing live broadcast operation can start the live client through the terminal and further use the live function.

The current game picture refers to a game picture of the target game operated by a main player user who live broadcasts the target game through a live broadcast client. Specifically, the manner of obtaining the current game picture may include multiple manners, for example, the current game picture may be obtained from a live video stream, or the current game picture may be obtained from a live interface of a live client, or the current game picture may be obtained from a game program run by the terminal.

In the embodiment of the application, the anchor user can live the target game through various devices, for example, a computer or a mobile phone.

For example, when a game is played directly through a computer, a user of the anchor firstly runs a game program through a mobile phone, and a game picture is displayed on a mobile phone interface (the game picture displayed on the mobile phone interface can be obtained by rendering game data); then, running a live program through a computer, and acquiring an image video of a main broadcasting user through a camera on the computer; the game picture of the mobile phone interface is projected to a computer, and the main broadcast image video (if the main broadcast image video exists) and the game picture video are spliced to form a live broadcast picture displayed by a computer screen; and sending the live broadcast picture to live broadcast ends of other audiences through a live broadcast server.

For another example, when the mobile phone is used for live broadcasting, the mobile phone of the anchor user runs a game program, and the mobile phone interface displays a game picture (the game picture displayed on the mobile phone interface can be obtained by rendering game data); a live broadcast program is run in the background of the mobile phone, and a game picture is recorded; and sending the recorded game picture to the live broadcast end of other audiences through the live broadcast server.

Further, in this embodiment of the application, the game screen acquired through the game program may be a game screen obtained by rendering according to game data, or the game screen acquired through the live program may be a captured game screen.

102. And determining at least one current game element according to the content in the current game picture.

The current game element refers to different types of content displayed in the current game screen, for example, the content may be classified into dynamic content and static content according to the status attribute, the dynamic content may be each game scene in the current game screen, and the dynamic content may be a virtual game character in the current game screen.

In some embodiments, in order to improve the accuracy of the identification of the game screen, the step "determining at least one current game element according to the content in the current game screen" may include the following operations:

carrying out scene recognition on the current game picture, and determining the current game scene in the current game picture;

carrying out role identification on the current game picture, and determining the current virtual game role in the current game picture;

and obtaining the current game element based on the current game scene and the current virtual game role.

The scene recognition of the current game picture refers to recognition of static content in the current game picture, and the static content in the current game picture can be determined first, and then the scene recognition of the static content is performed.

The static content may be a game scene in the current game screen, that is, the current game scene, and thus the static content may be subjected to image recognition, and specifically, the static content may be subjected to recognition of multiple dimensions, such as shape, size, color, and the like, so that the current game scene included in the current game screen may be determined.

For example, an image of the static content of the current game screen is acquired, the shape, size, and color of the image are recognized, and objects such as an enclosure and a river are recognized, so that it can be determined that the current game scene included in the current game screen includes the enclosure and the river.

The character recognition of the current game picture refers to the recognition of dynamic content in the current game picture, and the dynamic content in the current game picture can be determined firstly, and then the character recognition of the dynamic content is carried out.

Since the dynamic content may be a virtual game character in the current game screen, that is, the current virtual game character, the character recognition may be performed on the dynamic content, and specifically, the current virtual game character may be recognized by matching the dynamic content with a preset virtual game character of the target game.

For example, if the dynamic content in the current game screen is acquired, the dynamic content is matched with a preset virtual game character, and the dynamic content is determined to conform to a first virtual game character, a second virtual game character, and the like of the preset virtual game character, it may be determined that the current virtual game character includes: a first virtual game character, a second virtual game character.

Further, the current game elements in the current game picture can be obtained according to the identified current game scene and the current virtual game role, for example, the current game scene can include an enclosure and a river; the current virtual game character includes a first virtual game character and a second virtual game character, then it may be determined that the current game element includes: an enclosure, a river, a first virtual game character and a second virtual game character.

In some embodiments, in order to improve the video recommendation efficiency, before the step "determining at least one current game element according to the content in the current game screen", the following steps may be further included:

acquiring voice information of a live broadcast client;

recognizing the voice information and determining a keyword of the voice information;

and if the keywords accord with the keywords recommended by the video, executing a step of determining at least one current game element according to the content in the current game picture.

The voice information comprises voice of a main player user who carries out live broadcast on the target game through a live broadcast client, or game audio in the target game.

And if the voice information is the voice of the anchor user, recognizing the voice of the anchor user, and extracting key words in the voice of the anchor user according to a recognition result.

When the voice information is the voice of the anchor user, the keywords recommended by the video may include "recommendation", "previous period", "current period", "watching", "video", and the like. If the extracted keywords in the voice of the anchor user accord with the keywords recommended by the video, the anchor user can be determined to recommend the current game live video. Further, the step of determining at least one current game element based on the content in the current game screen may be performed.

If the voice information is the game audio in the target game, that is, the anchor user is in the operation process of the target game, the auxiliary broadcast sound in the target game, for example, when the game in the opposite party's game formation is defeated is a player, the auxiliary broadcast sound is generated to be xxx to defeat xxx, so that the keyword in the game audio can be extracted.

When the voice information is game audio, the keywords recommended by the video may include "beat" and the like, and if the keywords in the game audio match the keywords recommended by the video, the step of determining at least one current game element according to the content in the current game picture may be performed.

In some embodiments, in order to improve the viewing experience of the user viewing the live video, the video recommendation method may further include the following steps:

if the keywords do not accord with the keywords recommended by the video, acquiring historical watching behavior information of the user;

and determining a target video from the historical live video based on the historical viewing behavior information.

The historical watching behavior information refers to an operation of a user on a terminal in a process of watching a live video through a live client of the terminal, and may include an operation of video search through the terminal.

Furthermore, the type of the video preferred by the user can be detected according to the historical watching behavior information of the user, and then the target video can be selected from the current live video by combining the video type and the content of the current game picture to recommend the user, so that the video which the user most wants to watch can be recommended.

In some embodiments, in order to determine an accurate game element from the current game screen, the step "determining at least one current game element according to the content in the current game screen" may further include the following operations:

and determining the current game element according to the keywords and the content in the current game picture.

In the embodiment of the application, the game elements can be determined according to the keywords of the voice of the anchor user and recommended.

For example, the capture of the anchor user's voice may be "i last live play xxx hero, well played in highland, kill how many people," then the game element may be determined to be: xxx hero, highland. Then, the following are recognized from the current game picture: xxx hero, highland, the current game element is available.

103. The target video is determined from historical live video of the target game through the current game element.

The historical live video of the target game comprises live game video of the current anchor user live target game before the current time and live game video of other anchor users live target games.

In some embodiments, the current game element at least comprises a current game scene in the current game picture and at least one virtual game character in the current game picture, and in order to improve the accuracy of video recommendation, the step "determining a target video from historical live video of a target game through the current game element and current game information" may include the following operations:

determining the scene type of the current game scene in a game map of the target game;

determining a target virtual game role operated by an anchor user from current virtual game roles;

and determining a target video from the historical live video according to the scene type and the target virtual game role.

In the embodiment of the application, the game map of the target game comprises a plurality of game scenes, wherein the same game scene exists at different positions in the game map, and in order to improve the accuracy of game scene identification, the scene type of the current game scene in the game map of the target game can be determined.

Furthermore, the current virtual game roles can be multiple, including the virtual game roles operated by the anchor user and other game players, so that the virtual game roles operated by the anchor user, namely the target virtual game roles, can be screened from the current virtual game roles in order to recommend the live game video of the anchor user.

In the game picture of the target game, the virtual game role is shown along with the name and the life bar of the operation user of the virtual game role, and the life bar indicates the life value of each virtual game role. The colors of the life bars displayed by the virtual game characters in different game arrangements on the game interface of the target game of the anchor user can be different, the color of the life bar of the virtual game character operated by the anchor user can be different from the color of the life bar of the virtual game character in the same game arrangement, and the target virtual game character operated by the anchor user can be determined from the virtual game characters according to the color of the life bar corresponding to the virtual game character in the current game picture.

For example, in the current game screen, a first virtual game character, a second virtual game character, and a third virtual game character are included, where a color of a life bar of the first virtual game character may be red, a color of a life bar of the second virtual game character may be blue, and a color of a life bar of the third virtual game character may be green, where a color of a life bar of a virtual game character operated by a game player displayed on a game interface of the game player in the target game may be green, it may be determined that the third virtual game character may be a target game character operated by the host user.

determining a target position of a current game scene in a game map of a target game;

and determining a target video from the historical live video according to the target position and the target virtual game role.

In the embodiment of the application, the game map of the target game comprises a plurality of game scenes, wherein the same game scenes are located at different positions in the game map, and in order to improve the accuracy of game scene identification, the scene position of the current game scene in the game map of the target game, namely the target position, can be determined. And then determining a target video from the historical live video according to the target position and the target virtual role.

In some embodiments, to improve the accuracy of video recommendation, the step "determining a target video from historical live videos according to scene type, target virtual game character" may include the following operations:

determining a target historical live video matched with the scene type, the target virtual game role and the current game process from the historical live video;

and determining a target video based on the target historical live video.

Specifically, the content of each historical live video is matched with the scene type, the target virtual game role and the current game process, and the historical live video similar to the scene type, the target virtual game role and the current game process at the same time is screened out from the multiple historical live videos to obtain the target historical live video.

For example, historical live video may include: the first historical live video, the second historical live video, the third historical live video, the fourth historical live video and the like. And respectively matching each historical live video with the scene type, the target virtual game role and the current game progress, and if the historical live video similar to the scene type, the target virtual game role and the current game progress is the first historical live video, determining that the first historical live video is the target historical live video.

In some embodiments, in order to more accurately recommend a live video that a viewer wants to watch, the step "determine a target video based on a target historical live video" may include the following operations:

acquiring at least one marked video clip in a target historical live video;

the target video is determined from the marked video segments.

The marked video clip refers to a certain video clip with a recommendation score meeting a preset video recommendation score in the historical live video. Because the duration of the whole historical live video is long, and the content of the historical live video comprises video clips which are not interested by the user, the video clips in the historical live video can be marked for the convenience of watching recommended videos by audiences.

Specifically, the target video is determined from the marked video segments, and the video segment with the highest similarity to the scene type, the target virtual game role and the current game process can be selected from the marked video, so that the target video can be obtained.

In some embodiments, to improve video recommendation efficiency, before the step "obtaining at least one marked video segment in the target historical live video", the following steps may be further included:

acquiring a historical game picture when game battle performance information of a main broadcasting user in a target historical live video accords with preset battle performance information;

the marked video segments are generated based on the historical game elements in the historical game screens and the historical game progress.

The method comprises the steps of firstly, obtaining game battle performance information of a main broadcasting user in a target historical live video when the main broadcasting user is in different game progresses in a target game, further, determining the game battle performance information which accords with preset battle performance information from the target historical live video, and then obtaining a game picture which corresponds to the game battle performance information which accords with the preset battle performance information to obtain a historical game picture.

The preset battle performance information is used for evaluating the performance of the game battle performance of the anchor user, and when the game battle performance information accords with the preset battle performance information, the game battle performance of the anchor user can be represented to be better; when the game battle performance information does not accord with the preset battle performance information, the poor game battle performance of the anchor user can be represented.

Further, identifying the historical game picture to obtain historical game elements (namely game scenes and virtual game roles) in the historical game picture, acquiring game data according to the content in the historical game picture to obtain a historical game process corresponding to the historical game picture, finally intercepting video clips in the historical live video according to the historical game elements and the historical game process to obtain marked video clips, and generating mark information of the marked video clips so as to improve the matching efficiency of the video clips when the historical live video is matched.

In some embodiments, to facilitate recommendation of the anchor user's historical live video, the step "determining a target video from the historical live video of the target game through the current game element" may include the following operations:

determining a live video corresponding to a main broadcast user from the historical live video to obtain a candidate live video;

and determining a target video from the candidate live videos through the current game element.

Specifically, all historical live videos corresponding to the target game are obtained, and then live videos corresponding to the anchor user are screened out from all the historical live videos, namely the live videos which are live by the anchor user, so that candidate live videos can be obtained.

For example, historical live video may include: the first historical live broadcast video, the second historical live broadcast video, the third historical live broadcast video, the fourth historical live broadcast video and the like, wherein the first historical live broadcast video and the second historical live broadcast video are live broadcast videos which are live broadcast by a main broadcast user, and the first historical live broadcast video and the second historical live broadcast video can be determined to be candidate live broadcast videos.

Further, a target video is determined from the candidate live videos through the current game element and the current game process, and the specific implementation of this step may refer to the above description, which is not repeated herein.

In some embodiments, to expand the video recommendation scope, the step "determine a target video from a historical live video of the target game by a current game element and current game information" may include the following operations:

determining live broadcast videos corresponding to target virtual game roles from historical live broadcast videos to obtain candidate live broadcast videos;

Specifically, all historical live videos corresponding to the target game are obtained, and then live videos including the target virtual game role are screened out from all the historical live videos, so that candidate live videos can be obtained.

For example, historical live video may include: the game device comprises a first historical live video, a second historical live video, a third historical live video, a fourth historical live video and the like, wherein the first historical live video and the second historical live video are live videos comprising target virtual game roles, and the first historical live video and the second historical live video can be determined to be candidate live videos.

In some embodiments, to improve the video recommendation accuracy, before the step "determining at least one current game element according to the content in the current game screen", and after the step "displaying the target video on the live interface of the live client", the following steps may be further included:

determining current game information of the target game according to the content in the current game picture;

and determining the target video from the historical live video of the target game through the current game element and the current game information.

In some embodiments, in order to improve the accuracy of the identification of the game screen, the step "determining the current game information of the target game according to the content in the current game screen" may include the following operations:

extracting game battle performance information and game play duration information from the content of the current game picture;

and determining the game progress based on the game battle performance information and the game-play duration information to obtain the current game information.

Wherein, the game battle performance information can refer to the game battle performance of the host user in the target game currently.

For example, the game performance may include the number of defeaters, and the like, and the higher the number of defeaters, the higher the game performance, and correspondingly, the higher the recommendation index of the live video including the current game picture.

Wherein, the game play duration refers to the duration from the start of the target game to the current time of the anchor user. Further, the game data difference between game battles in the current target game, such as an economic difference value, a grade difference value and the like, can be obtained.

For example, the target game may be a multiplayer battle game, and may be divided into two game camps including a first game camps and a second game camps, with game players between the two game camps playing the battle. The anchor user can be a game player of the first game battle, the game battle performance of the anchor user, the game data difference between the first game battle and the second game battle and the game match duration are obtained, the game fierce degree of the current game picture can be determined according to the game data difference of the game battle performance and the game match duration, and the game progress can be determined according to the game fierce degree.

104. And displaying the target video on a live interface of the live client.

After the target video is determined, video recommendation can be performed on the user of the live client according to the target video.

In some embodiments, to implement video recommendation to a user, the step "display target video on a live interface of a live client" may include the following operations:

and sending the target video to the live client so that the live client can display the target video on a game live interface of the current game picture.

Specifically, when a live interface of the live client displays a current game picture, the target video is sent to the live client, so that the live client can display the target video and the current game picture on the live interface together. So that the user can watch the target video quickly.

In some embodiments, the number of the target videos may be multiple, and in order to meet the video viewing requirement of the user, the step "displaying the target videos on a live interface of a live client" may include the following operations:

calculating the similarity of each target video and the content of the current game picture;

and sending a plurality of target videos to the live client so that the live client can display the target videos in a game live interface in an arranging manner according to the similarity.

Specifically, the video content of the target video may be acquired, and then the similarity between the video content of the target video and the content of the current game screen may be calculated. And sending a plurality of target videos and the similarity between each target video and the content of the current game picture to the live client, so that the live client can arrange and display each target video on a live interface according to the similarity between each target video and the content of the current game picture, for example, the target videos can be arranged in sequence from high to low according to the similarity. And the user of the live client can select the target video to watch according to the requirement.

In the embodiment of the application, the game content in the live video stream in which the anchor user live broadcasts the target game is identified in multiple dimensions, for example, the identification of a game map scene, the identification of the output voice of the anchor user and the data of a game process can be included, and highlight moments in the live video are accurately marked by the anchor output voice and the data of the game process according to the game map scene. And then in subsequent live video live broadcast of real-time game, the game video clips of similar game scenes can be matched in real time, similar highlight moments can be recommended for the user, and the highlight moments can be directly positioned to the highlight moments of the current video, so that the operation cost of redundant searching and finding of the user is reduced.

The embodiment of the application discloses a video recommendation method, which comprises the following steps: acquiring a current game picture of a target game, and live broadcasting the target game through a live broadcasting client; determining at least one current game element according to the content in the current game picture, and determining current game information of the target game according to the content in the current game picture; determining a target video from historical live videos of the target game through current game elements and current game information; and displaying the target video on a live interface of the live client. According to the method and the device, the identification of multiple dimensions is carried out on the content in the game live video stream, the identification result is obtained through the data including the game scene, the game process and the like, the target video similar to the content of the current live video is selected from the current game live video according to the identification result, then video recommendation is carried out on the user of the live client according to the target video, the user can watch related videos quickly, and therefore video recommendation efficiency can be improved.

Based on the above description, the video recommendation method of the present application will be further described below by way of example. Referring to fig. 2, fig. 2 is a schematic flow chart of another video recommendation method according to an embodiment of the present application, and taking an example that the video recommendation method is applied to a server, a specific flow may be as follows:

201. the server obtains game pictures of a target game which is live broadcast by the main broadcast user through the live broadcast client in the historical time period.

In the embodiment of the application, the method for acquiring the game picture of the live target game of the anchor user in the historical time period comprises various modes, for example, the game picture can be acquired from a live video stream in real time when the anchor user live the target game in the historical time period; or, the game picture can be collected in the live video after the live target game of the anchor user is completed.

202. The server identifies the game picture to obtain the game scene and the virtual game role in the game picture.

Specifically, after the server acquires the game screen, the server may perform recognition and determination on the game scene in the game screen, so as to complete recognition of the game type, and further determine the type of game battle occurring in the game screen by positioning the place where the game scene in the game screen is located in the game map.

In the embodiment of the present application, the target game may be a multiplayer Online tactical competitive game, that is, a moba (multiplayer Online game arena) game, and the playing method of the game is that equipment needs to be purchased in Battle, players are generally divided into two teams, the two teams compete with each other in a scattered game map, and each player controls a game role selected by the player.

For example, important battles often occur in some important game scenes of the target game, and the key shapes, areas and differentiated colors in the game map terrain included in the game scenes can be identified by performing image recognition on the content of the game scenes, for example, one important game scene of the target game may include key elements such as semi-circular enclosing walls, enclosing walls at high ground junctions, white tiles different from grass in high ground, and the identified game scene may be determined based on the identification of the key elements in the image recognition process.

Specifically, the image recognition of the fixed game scene in the game map of the target game can recognize the area ratio, the shape and the color of a river, the trend, the area size, the shape and the like of a fence, so that the game scene can be quickly and accurately recognized as the fixed game scene, the game scene in the live video needing to be compared is subjected to similarity matching with the fixed game scene, and if the matching degree is high, the same game scene can be judged.

Specifically, the image recognition is carried out on the game picture, and the game scene in which battles occur in the game picture is positioned, so that the superposition recognition and judgment of the map position of the game scene of the video live broadcast stream can be completed in subsequent similar game live broadcast videos.

Furthermore, the virtual game role in the game picture can be identified, and the virtual game role operated by the anchor user in the target game is identified, so that the identification and judgment of the virtual game role superposition of the live video stream in subsequent similar live game videos can be completed.

For example, when a plurality of virtual game characters exist in the game picture, that is, when a plurality of people identify the game scene on the same page, the green life value can be positioned by identifying the display color of the life value of the virtual game character, so that the virtual game character below the green life value is determined to be the own virtual game character, that is, the virtual game character operated by the anchor user; another role identification method may also be: in the whole live video identification, the virtual game character with the longest occurrence time and the highest frequency can be judged as the virtual game character operated by the anchor user.

203. The server acquires game match data in the game picture and determines the recommendation index of the game picture according to the game match data.

The game fighting data may include battle performance information of the anchor user in the target game when the anchor user performs the target game, and a difference of game resources, such as an economic difference value and a level difference value, between a game formation corresponding to a virtual game character operated by the anchor user and a game formation of an opposite party.

Further, the game severity of the game picture and the game achievement of the anchor user are determined according to the game fighting data. The severity of the two-party game battle in the game picture can be comprehensively judged through the economic difference and the game duration of the two-party game battle, for example, the smaller the economic difference of the two-party game battle and the longer the game duration of the two-party game battle, the higher the severity can be represented, and the recommended value of the game video, namely the recommended index is higher.

In some embodiments, the game fighting data can also assist in image recognition, game character recognition and the like of the game picture, so that the recognition result of each dimension is more accurate.

204. The server generates tag information of a game screen based on the game scene, the virtual game character, and the recommendation index.

Wherein the marking information includes: game scene position, virtual game role/virtual game role type, recommendation index.

For example, the game scene may be a highland area in a game map, the virtual game character type may be an auxiliary game character, the recommendation index may be a score, e.g., the recommendation index may be 80 points, and the like, and the flag information for generating the game screen according to the game scene, the virtual game character, and the recommendation index may be: high land area, auxiliary game character, 80 points.

Based on the mark information, when a follow-up anchor user live broadcasts a target game, a video clip with high matching degree can be quickly matched from a historical live broadcast video, so that recommendation is carried out on audience users.

205. The server acquires a current game picture of a target game which is live broadcast by a current anchor user through a live broadcast client.

When the current anchor user live broadcasts the target game through the live broadcast client, the server can acquire the current game picture in the live broadcast video stream in real time.

In the embodiment of the present application, triggering to acquire the game video associated with the current game screen may include multiple ways. For example, the first triggering method may be: when the anchor user directly broadcasts the target game, the voice of the anchor user is synchronously recognized in real time, the semantic meaning or the trigger word of the voice of the anchor user is analyzed, and when the semantic meaning of the voice of the anchor user is the words of ' recommending ', ' previous period ', ' forward period ', ' last time ', ' before ' I ', ' watching ' and ' video ', the analysis can be carried out to obtain the game live broadcast video of which the anchor user wants to recommend the forward period.

When the live game video of the current period which the anchor user wants to recommend is identified, whether the aspects of the game scene, the virtual game role, the game process and the like of the current game picture are overlapped with the live game video of the current period can be identified, if the overlapping rate is high, the live game video of the current period can be recommended to the audience users of the live client according to the similarity priority.

Specifically, the second triggering method may be: when the voice of the anchor user does not have the trigger word, the scene coincidence degree of the current game picture and the game picture in the historical live video can be judged, if the scene coincidence degree is high, the situation that the scene coincidence degree is recommended to audience users of the client side can be considered, and the audience users can click to enter the video for checking the current highlight moment in the interest.

206. And the server screens out a target video from the historical live broadcast video according to the marking information and the content of the current game picture, and recommends audience users of the live broadcast client based on the target video.

Specifically, after the current game picture is obtained, the current game picture may be identified, a game scene in the current game picture is identified, a game character is virtualized, and then a game process corresponding to the current game picture is obtained.

The game scene, the virtual game role and the game process are matched with the mark information in the historical live video, the mark information with the highest matching degree is screened out, and the game video segment corresponding to the mark information in the historical live video is further determined, so that the target video can be obtained. The target video may then be recommended to the audience user of the live client.

In the embodiment of the application, highlight moments in a live video stream are recorded, and fine marking is performed through game data such as game places, used virtual game roles and game processes. So that each small node of each video segment has a refined label. By means of fine and tagged management of live video resources, each section of live video has the opportunity of being watched repeatedly and repeatedly, and relevance of video assets of a live platform is improved.

Furthermore, the tendency of video recommendation to audience users by the anchor user is identified through the voice keywords of the anchor user, or the video recommendation can be performed through the coincidence degree of two sections of live videos, if the coincidence degree is high, for example, the same virtual game role is used, the same type of virtual game role is used, the game progresses are consistent in similar game scene places (two parties are close in economy and the game-playing time is long), and if the coincidence degree reaches a certain set threshold value, the video can be automatically recommended to the audience users at the live client. Therefore, the video searching problem of the audience users can be solved, and when the anchor user mentions 'turn over the video before me' in the live broadcasting process, the audience users do not need to exit the live broadcasting room to search for the live broadcasting video on other video playing platforms, but can directly recommend the relevant video to the audience users, manual searching of the audience users is not needed, and user experience is improved.

The embodiment of the application discloses a video recommendation method, which comprises the following steps: the method comprises the steps that a server obtains a game picture of a live broadcast target game of a main broadcast user through a live broadcast client in a historical time period, identifies the game picture, obtains a game scene and virtual game characters in the game picture, obtains game fighting data in the game picture, determines a recommendation index of the game picture according to the game fighting data, generates marking information of the game picture based on the game scene, the virtual game characters and the recommendation index, obtains a current game picture of the current main broadcast user through the live broadcast client in a live broadcast mode, screens out a target video from historical live broadcast videos according to the marking information and the content of the current game picture, and recommends audience users of the live broadcast client based on the target video. Therefore, the accuracy of video recommendation of audience users of the live broadcast client can be improved.

In order to better implement the video recommendation method provided by the embodiment of the present application, the embodiment of the present application further provides a video recommendation device based on the video recommendation method. The meaning of the noun is the same as that in the video recommendation method, and specific implementation details can refer to the description in the method embodiment.

Referring to fig. 3, fig. 3 is a block diagram of a video recommendation apparatus according to an embodiment of the present disclosure. The device includes:

a first obtaining unit 301, configured to obtain a current game screen of a target game, where the target game is live broadcast through a live broadcast client;

a first determining unit 302, configured to determine at least one current game element according to content in the current game screen;

a second determining unit 303, configured to determine a target video from historical live videos of the target game through the current game element;

and the display unit 304 is configured to display the target video on a live interface of the live client.

In some embodiments, the second determining unit 303 may include:

In some embodiments, the third determining subunit may be specifically configured to:

determining the target video based on the target historical live video.

acquiring at least one marked video clip in the target historical live video;

determining the target video from the marked video segments.

In some embodiments, the third determining subunit may be further specifically configured to:

In some embodiments, the second determining unit 303 may include:

In some embodiments, the apparatus may further comprise:

In some embodiments, the third determining unit may include:

In some embodiments, the first determining unit 302 may include:

In some embodiments, the apparatus may further comprise:

In some embodiments, the first determining unit 302 may include:

In some embodiments, the display unit 304 may include:

The embodiment of the application discloses a video playing device, wherein a first obtaining unit 301 is used for obtaining a current game picture of a target game, and the target game is live broadcast through a live broadcast client; the first determining unit 302 determines at least one current game element according to the content in the current game picture; the second determining unit 303 determines a target video from the historical live video of the target game through the current game element; the display unit 304 displays the target video on a live interface of the live client. Thus, video recommendation efficiency can be improved.

Correspondingly, the embodiment of the application also provides a computer device, and the computer device can be a server. As shown in fig. 4, fig. 4 is a schematic structural diagram of a computer device according to an embodiment of the present application. The computer apparatus 400 includes a processor 401 having one or more processing cores, a memory 402 having one or more computer-readable storage media, and a computer program stored on the memory 402 and executable on the processor. The processor 401 is electrically connected to the memory 402. Those skilled in the art will appreciate that the computer device configurations illustrated in the figures are not meant to be limiting of computer devices and may include more or fewer components than those illustrated, or some components may be combined, or a different arrangement of components.

The processor 401 is a control center of the computer device 400, connects the respective parts of the entire computer device 400 using various interfaces and lines, performs various functions of the computer device 400 and processes data by running or loading software programs and/or modules stored in the memory 402 and calling data stored in the memory 402, thereby monitoring the computer device 400 as a whole.

In the embodiment of the present application, the processor 401 in the computer device 400 loads instructions corresponding to processes of one or more application programs into the memory 402 according to the following steps, and the processor 401 runs the application programs stored in the memory 402, thereby implementing various functions:

acquiring a current game picture of a target game, and live broadcasting the target game through a live broadcasting client; determining at least one current game element according to the content in the current game picture; determining a target video from historical live videos of the target game through a current game element; and displaying the target video on a live interface of the live client.

The above operations can be implemented in the foregoing embodiments, and are not described in detail herein.

Optionally, as shown in fig. 4, the computer device 400 further includes: touch-sensitive display screen 403, radio frequency circuit 404, audio circuit 405, input unit 406 and power 407. The processor 401 is electrically connected to the touch display screen 403, the radio frequency circuit 404, the audio circuit 405, the input unit 406, and the power source 407. Those skilled in the art will appreciate that the computer device configuration illustrated in FIG. 4 does not constitute a limitation of computer devices, and may include more or fewer components than those illustrated, or some components may be combined, or a different arrangement of components.

The touch display screen 403 may be used for displaying a graphical user interface and receiving operation instructions generated by a user acting on the graphical user interface. The touch display screen 403 may include a display panel and a touch panel. The display panel may be used, among other things, to display information entered by or provided to a user and various graphical user interfaces of the computer device, which may be made up of graphics, text, icons, video, and any combination thereof. Alternatively, the Display panel may be configured in the form of a Liquid Crystal Display (LCD), an Organic Light-Emitting Diode (OLED), or the like. The touch panel may be used to collect touch operations of a user on or near the touch panel (for example, operations of the user on or near the touch panel using any suitable object or accessory such as a finger, a stylus pen, and the like), and generate corresponding operation instructions, and the operation instructions execute corresponding programs. Alternatively, the touch panel may include two parts, a touch detection device and a touch controller. The touch detection device detects the touch direction of a user, detects a signal brought by touch operation and transmits the signal to the touch controller; the touch controller receives touch information from the touch sensing device, converts the touch information into touch point coordinates, sends the touch point coordinates to the processor 401, and can receive and execute commands sent by the processor 401. The touch panel may overlay the display panel, and when the touch panel detects a touch operation thereon or nearby, the touch panel may transmit the touch operation to the processor 401 to determine the type of the touch event, and then the processor 401 may provide a corresponding visual output on the display panel according to the type of the touch event. In the embodiment of the present application, the touch panel and the display panel may be integrated into the touch display screen 403 to realize input and output functions. However, in some embodiments, the touch panel and the touch panel can be implemented as two separate components to perform the input and output functions. That is, the touch display screen 403 may also be used as a part of the input unit 406 to implement an input function.

In the embodiment of the present application, a game application is executed by the processor 401 to generate a graphical user interface on the touch display screen 403, where a virtual scene on the graphical user interface includes at least one skill control area, and the skill control area includes at least one skill control. The touch display screen 403 is used for presenting a graphical user interface and receiving an operation instruction generated by a user acting on the graphical user interface.

The rf circuit 404 may be used for transceiving rf signals to establish wireless communication with a network device or other computer device via wireless communication, and for transceiving signals with the network device or other computer device.

The audio circuit 405 may be used to provide an audio interface between a user and a computer device through speakers, microphones. The audio circuit 405 may transmit the electrical signal converted from the received audio data to a speaker, and convert the electrical signal into a sound signal for output; on the other hand, the microphone converts the collected sound signal into an electrical signal, which is received by the audio circuit 405 and converted into audio data, which is then processed by the audio data output processor 401, and then sent to, for example, another computer device via the radio frequency circuit 404, or output to the memory 402 for further processing. The audio circuit 405 may also include an earbud jack to provide communication of a peripheral headset with the computer device.

The input unit 406 may be used to receive input numbers, character information, or user characteristic information (e.g., fingerprint, iris, facial information, etc.), and to generate keyboard, mouse, joystick, optical, or trackball signal inputs related to user settings and function control.

The power supply 407 is used to power the various components of the computer device 400. Optionally, the power source 407 may be logically connected to the processor 401 through a power management system, so as to implement functions of managing charging, discharging, power consumption management, and the like through the power management system. The power supply 407 may also include one or more dc or ac power sources, recharging systems, power failure detection circuitry, power converters or inverters, power status indicators, or any other component.

Although not shown in fig. 4, the computer device 400 may further include a camera, a sensor, a wireless fidelity module, a bluetooth module, etc., which are not described in detail herein.

In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.

As can be seen from the above, the computer device provided in this embodiment obtains the current game picture of the target game, and the target game is live broadcast through the live broadcast client; determining at least one current game element according to the content in the current game picture; determining a target video from historical live videos of the target game through a current game element; and displaying the target video on a live interface of the live client.

It will be understood by those skilled in the art that all or part of the steps of the methods of the above embodiments may be performed by instructions or by associated hardware controlled by the instructions, which may be stored in a computer readable storage medium and loaded and executed by a processor.

To this end, the present application provides a computer-readable storage medium, in which a plurality of computer programs are stored, and the computer programs can be loaded by a processor to execute the steps in any of the video recommendation methods provided by the embodiments of the present application. For example, the computer program may perform the steps of:

and displaying the target video on a live interface of the live client.

Wherein the storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.

Since the computer program stored in the storage medium can execute the steps in any video recommendation method provided in the embodiments of the present application, beneficial effects that can be achieved by any video recommendation method provided in the embodiments of the present application can be achieved, which are detailed in the foregoing embodiments and will not be described herein again.

The video recommendation method, apparatus, storage medium, and computer device provided in the embodiments of the present application are described in detail above, and specific examples are applied herein to illustrate the principles and implementations of the present application, and the description of the embodiments above is only used to help understand the method and core ideas of the present application; meanwhile, for those skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.

Claims

1. A method for video recommendation, the method comprising:

acquiring a current game picture of a target game, wherein the target game is live broadcast through a live broadcast client;

determining a target video from historical live videos of the target game through the current game element;

and displaying the target video on a live interface of the live client.

2. The method of claim 1, wherein the current game element comprises at least a current game scene in the current game screen and at least a current virtual game character in the current game screen;

the determining a target video from historical live videos of the target game through the current game element comprises:

determining a scene type of the current game scene in a game map of the target game;

determining a target virtual game role operated by an anchor user from the current virtual game roles, wherein the anchor user is a user for live broadcasting a target game through the live broadcasting client;

3. The method of claim 2, wherein determining a target video from the historical live video based on the scene type and the target virtual game character comprises:

determining the target video based on the target historical live video.

4. The method of claim 3, wherein the determining the target video based on the target historical live video comprises:

acquiring at least one marked video clip in the target historical live video;

determining the target video from the marked video segments.

5. The method of claim 4, further comprising, prior to said obtaining at least one tagged video segment of said target historical live video:

6. The method of claim 2, wherein the determining a target video from historical live video of the target game via the current game element comprises:

determining a live video corresponding to the anchor user from the historical live video to obtain a candidate live video;

determining the target video from the candidate live videos through the current game element.

7. The method of claim 2, wherein the determining a target video from historical live video of the target game via the current game element comprises:

determining live broadcast videos corresponding to the target virtual game role from the historical live broadcast videos to obtain candidate live broadcast videos;

8. The method of claim 2, wherein the determining a target video from historical live video of the target game via the current game element comprises:

determining a target position of the current game scene in a game map of the target game;

determining a target virtual game role operated by an anchor user from the current virtual game roles;

9. The method of claim 1, further comprising, after displaying the target video on a live interface of the live client prior to the determining at least one current game element from content in the current game view:

and determining a target video from the historical live video of the target game through the current game element and the current game information.

10. The method of claim 9, wherein determining the current game information of the target game according to the content in the current game screen comprises:

11. The method of claim 1, wherein determining at least one current game element based on content in the current game screen comprises:

and obtaining a current game element based on the current game scene and the current virtual game role to obtain the current game element.

12. The method of claim 1, further comprising, prior to said determining at least one current game element based on content in the current game screen:

acquiring voice information of the live broadcast client;

and if the keywords accord with the keywords recommended by the video, executing the step of determining at least one current game element according to the content in the current game picture.

13. The method of claim 12, further comprising:

determining the target video from the historical live video based on the historical viewing behavior information.

14. The method of claim 12, wherein determining at least one current game element based on content in the current game screen comprises:

and determining the current game element according to the keyword and the content in the current game picture.

15. The method of claim 1, wherein the displaying the target video on a live interface of the live client comprises:

and sending the target video to the live client so that the live client displays the target video on a game live interface of the current game picture.

16. A video recommendation apparatus, characterized in that the apparatus comprises:

17. A computer device comprising a memory, a processor and a computer program stored on the memory and running on the processor, wherein the processor implements the video recommendation method of any of claims 1-15 when executing the program.

18. A storage medium storing instructions adapted to be loaded by a processor to perform the video recommendation method of any of claims 1 to 15.