Video bullet screen display method and device, equipment and medium

Abstract

The invention provides a video bullet screen display method and device, equipment and a medium, and relates to the technical field of video processing, in particular to the technical field of artificial intelligence and video playing. According to the implementation scheme, multiple first bullet screens for a target video clip are acquired; based on the text content of each first bullet screen in the plurality of first bullet screens, determining a first score of the first bullet screen, the first score of each first bullet screen being determined by predicting the preference of a user for the first bullet screen; based on the video content of the target video clip, determining a second score of the target video clip, the second score being determined by predicting the preference of the user to the video content; determining a target number based on the second score; in response to determining that the number of the plurality of first bullet screens exceeds the target number, determining a target number of target bullet screens from the plurality of first bullet screens based on the first score of each first bullet screen; and displaying the target number of target bullet screens when the target video clip is played.

Description

Technical Field

[0001] The present disclosure relates to the field of video processing technology, in particular to the field of artificial intelligence and video playback technology, and specifically to a video bullet screen display method, device, electronic device, computer-readable storage medium and computer program product. Background Art

[0002] Artificial intelligence is a discipline that studies how to use computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It includes both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing; artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, as well as machine learning / deep learning, big data processing technology, knowledge graph technology, and other major directions.

[0003] Video is an important information medium with rich media features that can provide an immersive experience. With the development of information technology, it is becoming more and more popular among users. Video barrage is a comment sent by users in real time while watching a video, usually displayed on the video screen in the form of scrolling text. Viewers can post barrages while watching a video to express their opinions in real time, thereby enabling interaction between users and improving the video viewing experience.

[0004] The methods described in this section are not necessarily methods that have been previously conceived or employed. Unless otherwise indicated, it should not be assumed that any method described in this section is considered to be prior art simply because it is included in this section. Similarly, unless otherwise indicated, the issues mentioned in this section should not be considered to have been recognized in any prior art. Summary of the invention

[0005] The present disclosure provides a video barrage display method, device, electronic device, computer-readable storage medium and computer program product.

[0006] According to one aspect of the present disclosure, a method for displaying video barrages is provided, comprising: obtaining multiple first barrages for a target video clip; determining a first score of each first barrage based on the text content of the multiple first barrages, wherein the first score of each first barrage is determined by predicting a user's preference for the first barrage; determining a second score of the target video clip based on the video content of the target video clip, wherein the second score is determined by predicting a user's preference for the video content; determining a target number based on the second score; in response to determining that the number of the multiple first barrages exceeds the target number, determining a target number of target barrages from the multiple first barrages based on the first score of each first barrage; and displaying the target number of target barrages when playing the target video clip.

[0007] According to another aspect of the present disclosure, a video barrage display device is provided, including: an acquisition unit, configured to acquire multiple first barrages for a target video clip; a first determination unit, configured to determine a first score of each first barrage in the multiple first barrages based on the text content of the first barrage, wherein the first score of each first barrage is determined by predicting a user's preference for the first barrage; a second determination unit, configured to determine a second score of the target video clip based on the video content of the target video clip, wherein the second score is determined by predicting a user's preference for the video content; a third determination unit, configured to determine a target number based on the second score; a fourth determination unit, configured to determine a target number of target barrages from the multiple first barrages based on the first score of each first barrage in response to determining that the number of the multiple first barrages exceeds the target number; and a display unit, configured to display the target number of target barrages when playing the target video clip.

[0008] According to another aspect of the present disclosure, an electronic device is provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned video barrage display method.

[0009] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable the computer to execute the above-mentioned video barrage display method.

[0010] According to another aspect of the present disclosure, a computer program product is provided, including a computer program, wherein the computer program can implement the above-mentioned video barrage display method when executed by a processor.

[0011] According to one or more embodiments of the present disclosure, the barrage display during video playback can be optimized to enhance the user's video viewing experience.

[0012] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The accompanying drawings exemplarily illustrate the embodiments and constitute a part of the specification, and together with the text description of the specification, are used to explain the exemplary implementation of the embodiments. The embodiments shown are for illustrative purposes only and do not limit the scope of the claims. In all drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0014] Figure 1 A schematic diagram showing an exemplary system in which the various methods described herein may be implemented according to an exemplary embodiment of the present disclosure;

[0015] Figure 2 A flowchart of a video bullet screen display method according to an exemplary embodiment of the present disclosure is shown;

[0016] Figure 3A-3B A schematic diagram showing a process of processing a video bullet screen according to an exemplary embodiment of the present disclosure is shown;

[0017] Figure 4 A structural block diagram of a video bullet screen display device according to an exemplary embodiment of the present disclosure is shown;

[0018] Figure 5 A structural block diagram of an exemplary electronic device that can be used to implement the embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0019] The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope of the present disclosure. Similarly, for the sake of clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0020] In the present disclosure, unless otherwise specified, the use of the terms "first", "second", etc. to describe various elements is not intended to limit the positional relationship, timing relationship, or importance relationship of these elements, and such terms are only used to distinguish one element from another element. In some examples, the first element and the second element may refer to the same instance of the element, and in some cases, based on the description of the context, they may also refer to different instances.

[0021] The terms used in the description of various examples in this disclosure are only for the purpose of describing specific examples and are not intended to be limiting. Unless the context clearly indicates otherwise, if the number of elements is not specifically limited, the element can be one or more. In addition, the term "and / or" used in this disclosure covers any one of the listed items and all possible combinations.

[0022] The embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings.

[0023] Figure 1 FIG. 1 is a schematic diagram of an exemplary system 100 in which various methods and apparatuses described herein may be implemented according to an embodiment of the present disclosure. Figure 1 , the system 100 includes one or more client devices 101, 102, 103, 104, 105, and 106, a server 120, and one or more communication networks 110 coupling the one or more client devices to the server 120. The client devices 101, 102, 103, 104, 105, and 106 may be configured to execute one or more applications.

[0024] In an embodiment of the present disclosure, the server 120 may run one or more services or software applications that enable execution of the video bullet screen display method.

[0025] In some embodiments, server 120 may also provide other services or software applications, which may include non-virtualized environments and virtualized environments. In some embodiments, these services may be provided as web-based services or cloud services, such as provided to users of client devices 101, 102, 103, 104, 105, and / or 106 under a software as a service (SaaS) model.

[0026] exist Figure 1In the configuration shown, the server 120 may include one or more components that implement the functions performed by the server 120. These components may include software components, hardware components, or a combination thereof that can be executed by one or more processors. Users operating client devices 101, 102, 103, 104, 105, and / or 106 may in turn utilize one or more client applications to interact with the server 120 to utilize the services provided by these components. It should be understood that a variety of different system configurations are possible, which may differ from the system 100. Therefore, Figure 1 is one example of a system for implementing the various methods described herein and is not intended to be limiting.

[0027] The user may use client devices 101, 102, 103, 104, 105 and / or 106 to send the target video clip or bullet comment data to be processed. The client device may provide an interface that enables the user of the client device to interact with the client device. The client device may also output information to the user via the interface. Figure 1 Only six client devices are depicted, but one skilled in the art will appreciate that the present disclosure may support any number of client devices.

[0028] Client devices 101, 102, 103, 104, 105 and / or 106 may include various categories of computer devices, such as portable handheld devices, general-purpose computers (such as personal computers and laptop computers), workstation computers, wearable devices, smart screen devices, self-service terminal devices, service robots, game systems, thin clients, various messaging devices, sensors or other sensing devices, etc. These computer devices may run various categories and versions of software applications and operating systems, such as MICROSOFT Windows, APPLE iOS, UNIX-like operating systems, Linux or Linux-like operating systems (such as GOOGLE Chrome OS); or include various mobile operating systems, such as MICROSOFT Windows Mobile OS, iOS, Windows Phone, Android. Portable handheld devices may include cellular phones, smart phones, tablet computers, personal digital assistants (PDAs), etc. Wearable devices may include head-mounted displays (such as smart glasses) and other devices. Game systems may include various handheld game devices, Internet-enabled game devices, etc. The client device is capable of executing a variety of different applications, such as various Internet-related applications, communication applications (eg, email applications), short message service (SMS) applications, and may use various communication protocols.

[0029] The network 110 may be any type of network known to those skilled in the art that may support data communications using any of a variety of available protocols, including but not limited to TCP / IP, SNA, IPX, etc. By way of example only, one or more networks 110 may be a local area network (LAN), an Ethernet-based network, a token ring, a wide area network (WAN), the Internet, a virtual network, a virtual private network (VPN), an intranet, an extranet, a blockchain network, a public switched telephone network (PSTN), an infrared network, a wireless network (e.g., Bluetooth, WIFI), and / or any combination of these and / or other networks.

[0030] Server 120 may include one or more general purpose computers, dedicated server computers (e.g., PC (personal computer) servers, UNIX servers, mid-range servers), blade servers, mainframe computers, server clusters, or any other suitable arrangement and / or combination. Server 120 may include one or more virtual machines running virtual operating systems, or other computing architectures involving virtualization (e.g., one or more flexible pools of logical storage devices that may be virtualized to maintain a server's virtual storage device). In various embodiments, server 120 may run one or more services or software applications that provide the functionality described below.

[0031] The computing units in the server 120 may run one or more operating systems including any of the above operating systems and any commercially available server operating systems. The server 120 may also run any of a variety of additional server applications and / or middle-tier applications, including HTTP servers, FTP servers, CGI servers, JAVA servers, database servers, etc.

[0032] In some implementations, server 120 may include one or more applications to analyze and consolidate data feeds and / or event updates received from users of client devices 101, 102, 103, 104, 105, and 106. Server 120 may also include one or more applications to display data feeds and / or real-time events via one or more display devices of client devices 101, 102, 103, 104, 105, and 106.

[0033] In some embodiments, the server 120 may be a server of a distributed system, or a server combined with a blockchain. The server 120 may also be a cloud server, or an intelligent cloud computing server or intelligent cloud host with artificial intelligence technology. A cloud server is a host product in a cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and virtual private servers (VPS) services.

[0034] The system 100 may also include one or more databases 130. In some embodiments, these databases may be used to store data and other information. For example, one or more of the databases 130 may be used to store information such as audio files and video files. The databases 130 may reside in various locations. For example, the database used by the server 120 may be local to the server 120, or may be remote from the server 120 and may communicate with the server 120 via a network-based or dedicated connection. The databases 130 may be of different categories. In some embodiments, the databases used by the server 120 may be, for example, relational databases. One or more of these databases may store, update, and retrieve data to and from the databases in response to commands.

[0035] In some embodiments, one or more of the databases 130 may also be used by applications to store application data. The databases used by the applications may be different categories of databases, such as a key-value store, an object store, or a conventional store backed by a file system.

[0036] Figure 1 The system 100 may be configured and operated in various ways to enable the application of various methods and apparatuses described in the present disclosure.

[0037] Video is an important information medium, and it has the characteristics of rich media, which can provide an immersive experience. With the development of information technology, it is becoming more and more popular with users. Video barrage is the comment information sent by users in real time when watching videos, which is usually displayed on the video screen in the form of scrolling text. Viewers can express their opinions in real time by posting barrage while watching videos, so as to achieve interaction between users, and at the same time create a viewing atmosphere, provide supplementary information for video content, enhance the entertainment of watching videos, and effectively improve user experience. However, when the number of barrages is too large, the barrage display may cover the normal video content, thereby affecting the user's viewing of videos. In the related art, the number of barrage displays is usually limited based on a certain number threshold, or the user sets the barrage display range so that the barrage can only be displayed in the set range (such as the top of the video screen), and the number of barrage displays is limited based on this to avoid the barrage blocking the video content. However, this method will have a greater impact on the barrage content seen by users, affecting the interactive experience brought by the barrage to users.

[0038] Based on this, the present disclosure provides a method for displaying video barrages, which determines the target number of barrages corresponding to the video clip according to the score of the video clip content. When the current number of barrages exceeds the target number, the target number of barrages is screened out based on the score of the barrage, so that the number of barrages displayed is consistent with the importance of the video content, ensuring that the barrages bring a sense of interactivity to users while not affecting the display of the wonderful content of the video, thereby improving the user's viewing experience.

[0039] Figure 2 FIG. 2 shows a flow chart of a video bullet screen display method 200 according to an exemplary embodiment of the present disclosure. Figure 2 As shown, the method 200 includes:

[0040] Step S201, obtaining a plurality of first bullet comments for a target video segment;

[0041] Step S202: determining a first score of each first bullet screen in the plurality of first bullet screens based on the text content of the first bullet screen, wherein the first score of each first bullet screen is determined by predicting a user's preference for the first bullet screen;

[0042] Step S203: determining a second score of the target video segment based on the video content of the target video segment, wherein the second score is determined by predicting a user's preference for the video content;

[0043] Step S204: determining a target quantity based on the second score;

[0044] Step S205: in response to determining that the number of the plurality of first barrages exceeds the target number, determining a target number of target barrages from the plurality of first barrages based on the first score of each of the first barrages; and

[0045] Step S206: display the target number of target bullet screens when playing the target video clip.

[0046] By applying the above-mentioned video barrage display method 200, the first score can be used to indicate the importance of each first barrage (i.e., the degree to which it can attract users), and the second score can be used to indicate the excitement of the video content. Then, the target number of barrages corresponding to the video clip is determined according to the content score of the video clip. When the number of first barrages exceeds the target number, the target number of target barrages can be screened out based on the score of the barrage, so that the number of barrages displayed is consistent with the excitement of the video content, ensuring that the barrage brings a sense of interactivity to the user without affecting the display of the exciting content of the video, thereby enhancing the user's viewing experience.

[0047] In some examples, the target video segments can be obtained by segmenting the original video. For example, the original video can be segmented into multiple target video segments of equal length based on a certain target duration, or the original video can be segmented into multiple target video segments according to the number of barrages corresponding to the original video, so as to control the computational amount of barrage data processing for each target video segment, avoid repeated processing of the same barrage, and improve the efficiency of barrage processing.

[0048] According to some embodiments, the target video segment is obtained by: performing a shot analysis on the original video to determine a plurality of shots included in the original video; determining a plurality of video segments based on the plurality of shots; and determining the target video segment from the plurality of video segments. Thus, the original video can be segmented according to the shots, which facilitates the evaluation of the degree to which the video content of each shot can attract users, based on which the target number of bullet screen displays can be determined more accurately, thereby improving the efficiency of bullet screen data processing.

[0049] By performing physical shot analysis on the original video, the original video can be segmented into a series of temporally continuous shots, each shot being taken continuously by a camera. This segmentation is based on the physical properties of the video (such as the continuity of the video frame images). In some examples, physical shot analysis can be implemented using a transnet algorithm based on three-dimensional convolution. The convolution kernel of the three-dimensional convolution operation not only processes the video frame images in the spatial dimension, but also extracts the features of the video frame sequence in the dimension of the time axis, thereby more accurately identifying the shot boundaries in the video based on the spatial information and temporal information of the video.

[0050] According to some embodiments, obtaining multiple first bullet comments for the target video clip in step S201 includes: obtaining multiple initial bullet comments for the target video clip; and determining the multiple first bullet comments from the multiple initial bullet comments, wherein the semantic similarity between any two first bullet comments in the multiple first bullet comments is not higher than the similarity threshold. Thus, first bullet comments with different semantics can be screened from the multiple initial bullet comments, which is equivalent to deleting semantically repeated initial bullet comments, achieving semantic deduplication of bullet comments, avoiding a large number of semantically repeated bullet comments from swiping the video screen, and improving the user's viewing experience.

[0051] In some examples, the semantic similarity between each initial barrage can be determined using various methods. For example, the similarity of different barrages can be determined based on various indicators such as the text editing distance between the barrage texts of each initial barrage, the degree of difference between the probability distribution of the word frequency of the barrage texts, the proportion of simple common words, and the barrage deduplication can be achieved based on this.

[0052] According to some embodiments, determining the multiple first bullet comments from the multiple initial bullet comments includes: inputting the multiple initial bullet comments into a second language model to obtain multiple initial bullet comment sets output by the second language model, wherein the semantic similarity between the initial bullet comments in each initial bullet comment set is higher than the similarity threshold, and the second language model is obtained by fine-tuning the large language model using multiple sample initial bullet comments; and determining the multiple first bullet comments based on the multiple initial bullet comment sets. Thus, the language model can be used to perform semantic analysis on the initial bullet comments to divide the multiple initial bullet comments into multiple semantically similar initial bullet comment sets, based on which the first bullet comment screening result is obtained, and the natural language processing capability of the large-scale language model is used to improve the efficiency and accuracy of the first bullet comment screening.

[0053] In some examples, a large language model (LLM) is trained using a large-scale text dataset. On this basis, a sample corpus for fine-tuning training can be constructed using sample initial barrages, and reference output results can be annotated for the sample corpus so that the language model can output multiple first barrages with different semantics by analyzing the semantic correlation between multiple initial barrages.

[0054] According to some embodiments, method 200 further includes: for each target barrage in the target number of target barrages, determining the first number corresponding to the target barrage, wherein the first number indicates the number of initial barrages in the multiple initial barrages whose semantic similarity with the target barrage is higher than the similarity threshold; and based on the first number, determining the display parameters of the target barrage, wherein the display parameters include at least one of text color, text font size, and text transparency, and wherein, in step S406, displaying the target number of target barrages when playing the target video clip includes: when playing the target video clip, displaying the target barrage based on the display parameters of each target barrage. Thus, the number of initial barrages corresponding to each first barrage can be determined on the basis of semantic deduplication, and the barrage display effect can be determined based on this, so that the user can perceive the number of people who resonate with the content of the barrage based on the display effect of the barrage, thereby improving the user's viewing experience.

[0055] In some examples, multiple target barrages may be graded based on the first quantity corresponding to each target barrage, and the display effect of the target barrage may be determined according to the pre-configured display parameters corresponding to each target barrage level, so that the user can perceive the initial number of barrages corresponding to each target barrage, thereby improving the user's viewing experience.

[0056] In some examples, when multiple first bullet comments are determined based on multiple initial bullet comment sets output by the second language model described above, the first quantity corresponding to each first bullet comment can be determined based on the number of initial bullet comments in the initial bullet comment set corresponding to each first bullet comment. For example, structured information consisting of multiple first bullet comments and the first quantity corresponding to each first bullet comment can be obtained, so as to facilitate the direct acquisition of relevant information of multiple first bullet comments in the subsequent steps of determining the target bullet comment and displaying the target bullet comment.

[0057] As mentioned above, barrage comments are posted by video viewers while watching a video. Users can view barrage comments to receive other viewers' opinions while watching a video, thereby achieving interaction between users. In this case, barrage comments with monotonous and repetitive content or barrage comments that are irrelevant to the video content cannot arouse users' interest and cannot provide users with an interactive experience.

[0058] Based on this, according to some embodiments, determining the first score of each first bullet screen in the plurality of first bullet screens based on the text content of each first bullet screen in step S202 includes: for each first bullet screen in the plurality of first bullet screens, determining the first score of the first bullet screen based on at least one of the following items: the first relevance between the text content of the first bullet screen and the video content of the target video clip; and the value information of the first bullet screen, wherein the value information is determined based on the word frequency of at least one word included in the first bullet screen in the plurality of first bullet screens. Thus, the score of the bullet screen can be determined more accurately based on the heterogeneity between bullet screens and the relevance between the bullet screen and the video content.

[0059] It should be understood that the above method of determining the first score of the first bullet screen will not be affected by user characteristics, that is, the first score determined based on the above method will not change due to the change of the playback object of the target video segment. In this case, the first score of the first bullet screen can be predetermined and stored, thereby reducing the complexity of bullet screen data processing and reducing the occupation of online processing resources.

[0060] In some examples, the value information of the first barrage can be obtained using the TF-IDF (Term Frequency-Inverse Document Frequency) algorithm. The TF-IDF algorithm is a weighted calculation technology used for information retrieval and text mining, which can evaluate the importance of a word in a text library. The higher the TF-IDF value, the higher the importance of the word in the text library. In one example, the TF-IDF value can be determined based on the frequency of occurrence of the word in the text library. In one example, the TF-IDF value can be further determined in combination with the inverse document frequency of the word in the text library to more accurately indicate the importance of the word in the text library.

[0061] In some examples, the text library used to execute the TF-IDF algorithm can be a text library including the bullet text of multiple first bullets. In some examples, the text library can also include more content. For example, when multiple first bullets are determined from multiple initial bullets using the method described above, the text library can also include the bullet text of each initial bullet. In some examples, when the target video segment is obtained based on the original video, the text library can also include the bullet text of each initial bullet corresponding to the original video. In some examples, when the original video is an episode video in a video collection (such as a film and television series), the text library can also include the bullet text of each initial bullet corresponding to each episode video in the video collection. Based on this, the value information of each first bullet can be determined more comprehensively and accurately, and then the bullet rating can be determined more accurately, thereby improving the accuracy of the bullet display.

[0062] According to some embodiments, the first relevance between the text content of each first bullet comment and the video content of the target video segment is determined by: inputting the text content of the first bullet comment and one or more video frames of the target video segment into a relevance evaluation model to obtain the first relevance output by the relevance evaluation model, wherein the relevance evaluation model is trained using sample text content and sample video segments annotated with reference to the first relevance. Thus, the relevance evaluation model can be used to more efficiently and accurately determine the relevance between the bullet comment text and the video screen.

[0063] In some examples, the relevance evaluation model may be supervisedly trained using sample text content and sample video clips annotated with a reference first relevance to improve the accuracy of model prediction.

[0064] According to some embodiments, the relevance evaluation model includes a text encoding network and a video content encoding network, and the inputting the text content of the first bullet screen and one or more video frames of the target video segment into the relevance evaluation model to obtain the first relevance output by the relevance evaluation model includes: inputting the text content of the first bullet screen into the text encoding network; and inputting one or more video frames of the target video segment into the video content encoding network, so that the relevance evaluation model can determine the first relevance based on the encoding result of the text encoding network and the encoding result of the video content encoding network. Thus, the semantic space in which the bullet screen text and the video screen are mapped respectively by the text encoding network and the video content encoding network can be used to more accurately determine the first relevance based on the encoding result, that is, the double-tower correlation model including the text encoding network and the video content encoding network is used to more accurately determine the relevance between the text content and the video content.

[0065] In some examples, the first relevance between the text content of each first bullet screen and the video content of the target video clip can also be determined in other ways, such as by using a VLM model (Vision Language Model) to determine the content description text for the target video clip, and then determining the first relevance based on the relevance between the content description text and the bullet screen text content. In this way, the semantic parsing capability of the VLM model can be fully utilized to more easily and efficiently determine the first relevance between the text content of each first bullet screen and the video content of the target video clip.

[0066] According to some embodiments, in step S202, determining the first score of the first barrage based on the text content of each first barrage in the multiple first barrages includes: in response to determining that the multiple first barrages include a user barrage sent by a target user, for each first barrage in the multiple first barrages except the user barrage, based on the second correlation between the text content of the user barrage and the text content of the first barrage, determining the first score of the first barrage, and wherein, in step S406, displaying the target number of target barrages when playing the target video clip includes: displaying the target number of target barrages when playing the target video clip to the target user. Thus, it is possible to determine the importance of the barrage (i.e., the degree to which it can attract users) based on the content of the barrage itself, combined with the information of the target user, to determine whether each first barrage can resonate with the target user, thereby achieving user-level personalization and enhancing the interactive experience that the target barrage brings to the user.

[0067] According to some embodiments, the determining the first score of the first bullet screen based on the second relevance between the text content of the user bullet screen and the text content of the first bullet screen includes: inputting the text content of the user bullet screen and the text content of the first bullet screen into a first language model to obtain the second relevance output by the first language model, wherein the first language model is obtained by fine-tuning a large language model using sample text content annotated with a reference to the second relevance. Thus, the relevance of each first bullet screen to the user bullet screen can be determined more efficiently and accurately using a language model.

[0068] In some examples, the first language model and the aforementioned second language model may be the same or different. In some examples, the same language model may be used to analyze the semantic relevance between each initial bullet screen and the relevance between each first bullet screen and the user bullet screen, so as to improve the convenience of bullet screen data processing.

[0069] According to some embodiments, determining the second score of the target video segment based on the video content of the target video segment in step S203 includes: inputting the target video segment into a video evaluation model to obtain the second score output by the video evaluation model, wherein the video evaluation model is trained using sample video segments annotated with reference to the second score. Thus, the video content score can be determined more efficiently and accurately using the video evaluation model.

[0070] In some examples, the video evaluation model can be fine-tuned and trained on the basis of the visual language model to fully utilize the image semantic parsing ability of the visual language model and more efficiently determine the content score of the video clip.

[0071] In some examples, when the video frames of the target video clip include text content, the text content can be extracted based on the video frames, and the text extraction results and each video frame are input into the visual language model to obtain a more accurate video content score.

[0072] In some examples, when the target video clip includes audio content, ASR (Automatic Speech Recognition) technology can be used to recognize the audio content of the original video, and the audio text recognition results and each video frame can be input into the visual language model to obtain a more accurate video content score.

[0073] According to some embodiments, method 200 further includes: in response to determining that the multiple first bullet screens include a user bullet screen sent by a target user, determining the user bullet screen as a target bullet screen, and wherein in step S406, displaying the target number of target bullet screens when playing the target video clip includes: displaying the target number of target bullet screens when playing the target video clip to the target user. Thus, the user bullet screen of the target user can be directly determined as the target bullet screen, avoiding swallowing the bullet screen sent by the current target user and affecting user stickiness.

[0074] In some examples, determining the target number based on the second score in step S204 may be calculated based on a preset formula including a second score variable. In one example, the higher the second score of the target video clip, the smaller the corresponding target number, thereby preventing the barrage display from affecting the display of the video content, so that the user can immersively watch the wonderful content of the video, thereby improving the user's viewing experience.

[0075] In some examples, in response to determining that the number of the plurality of first bullet screens exceeds the target number in step S205, determining the target number of target bullet screens from the plurality of first bullet screens based on the first score of each first bullet screen may include: sorting based on the first score of each first bullet screen to obtain the target number of target bullet screens. Thus, when the number of targets displayed in the bullet screen is less than the number of first bullet screens, a target bullet screen that is more attractive to the user can be selected based on the bullet screen score, thereby improving the user's experience of watching the video.

[0076] Figure 3A-3B FIG. 1 is a schematic diagram showing a process of processing a video bullet screen according to an exemplary embodiment of the present disclosure. In this example, the target video segment can be obtained by segmenting the original video. Figure 3A As shown, the number of targets displayed in the barrage in the target video clip can be determined by the following steps:

[0077] Step S110, video evaluation. In this step, the target video segment can be input into the video evaluation model to obtain a second score output by the video evaluation model. The video evaluation model can evaluate the wonderfulness of the target video segment (i.e., the degree to which it can attract users), and determine a quantitative second score based on this.

[0078] Step S120, determining a target number. In this step, the target number of bullet screens to be displayed can be determined based on the second score of the target video clip, so that the number of bullet screens to be displayed is consistent with the excitement of the video content.

[0079] While determining the target number based on the video content of the target video clip, the processing of the video barrage may include the following steps:

[0080] Step S210, distinguish user bullet comments. In this step, for multiple initial bullet comments corresponding to the target video clip, determine whether each initial bullet comment is a user bullet comment sent by the target user. For user bullet comments, steps S220 and S230 can be skipped directly to determine it as the target bullet comment to be displayed. For non-user bullet comments, the target bullet comment can be selected from them through the following steps S220-S240.

[0081] Step S220, merging similar bullet comments. In this step, each non-user bullet comment can be input into the language model so that the language model can output multiple semantically similar initial bullet comment sets, and counting statistics are performed based on the output results of the language model to obtain multiple first bullet comments corresponding to the multiple initial bullet comment sets and the first quantity corresponding to each first bullet comment (i.e., the number of initial bullet comments included in the initial bullet comment set corresponding to each first bullet comment).

[0082] Step S230: Calculate the bullet comment score. This step can be implemented based on the following steps S31 to S34:

[0083] Step S231: Based on a text library including the barrage texts of all initial barrages corresponding to the original video, the inter-barrage heterogeneity score of each first barrage is determined using the TF-IDF algorithm.

[0084] Step S232: input the text content of each first barrage into the text coding network of the dual-tower correlation model as described above, and input one or more video frames of the target video clip into the video content coding network of the dual-tower correlation model, so that the dual-tower correlation model can determine the first correlation between the encoding results of the text coding network and the encoding results of the video content coding network based on the encoding results of the text coding network and the video content coding network, and then determine the picture correlation score of the first barrage.

[0085] Step S233: Determine the user relevance score of each first barrage based on the second relevance between the text content of each first barrage and the text content of the user barrage.

[0086] As mentioned above, although the user barrage in step S210 does not need to determine the target barrage through steps S220 and S230, since the text content of the user barrage is required in the scoring step for non-user barrages, the information of the user barrage obtained in step S210 still needs to be passed to step S230, and step S233 is executed based on this. In one example, multiple initial barrages may not include user barrages. In this case, the user relevance score of each first barrage can be set to a fixed value (for example, it can be set to zero) in step S233.

[0087] Step S234: weighted calculation is performed on the inter-barrage heterogeneity score, the screen relevance score and the user relevance score of each first barrage based on preset weights to obtain a first score for each first barrage.

[0088] Based on the determined first score of each first bullet screen and the target number of bullet screens, step S240 can be executed to determine the target bullet screen. In this step, the first score of each first bullet screen can be sorted to obtain the target number of target bullet screens.

[0089] As mentioned above, in step S240, the user bullet screen obtained in step S210 can be directly determined as the target bullet screen. In this case, the number of user bullet screens can be subtracted from the number of target screens to obtain the number of first bullet screens displayed. Based on this, the target bullet screen can be selected from multiple first bullet screens based on the first score of each first bullet screen and the number of first bullet screens displayed, and then the target bullet screen sent by the target user and the target bullet screen selected from bullet screens sent by other users can be displayed to the target user.

[0090] By executing the above-mentioned video barrage processing flow, the first score can be used to indicate the importance of each first barrage, and the second score can be used to indicate the excitement of the video content. Then, the number of barrage targets corresponding to the video clip can be determined according to the content score of the video clip, so that the number of barrage displays is consistent with the excitement of the video content, ensuring that the barrage brings a sense of interactivity to the user without affecting the display of the exciting content of the video, thereby improving the user's viewing experience.

[0091] According to one aspect of the present disclosure, a video barrage display device is also provided. Figure 4 FIG. 4 shows a structural block diagram of a video bullet screen display device 400 according to an exemplary embodiment of the present disclosure. Figure 4 As shown, the device 400 includes:

[0092] The acquisition unit 401 is configured to acquire a plurality of first bullet comments for a target video segment;

[0093] A first determining unit 402 is configured to determine a first score of each first bullet forum among the plurality of first bullet forums based on the text content of the first bullet forum, wherein the first score of each first bullet forum is determined by predicting a user's preference for the first bullet forum;

[0094] A second determining unit 403 is configured to determine a second score of the target video segment based on the video content of the target video segment, wherein the second score is determined by predicting a user's preference for the video content;

[0095] A third determining unit 404 is configured to determine a target quantity based on the second score;

[0096] A fourth determining unit 405 is configured to determine a target number of target barrages from the plurality of first barrages based on the first score of each of the first barrages in response to determining that the number of the plurality of first barrages exceeds the target number; and

[0097] The display unit 406 is configured to display the target number of target bullet comments when playing the target video clip.

[0098] According to some embodiments, the first determination unit 402 is configured to: determine, for each first barrage of the multiple first barrages, a first score of the first barrage based on at least one of the following items: a first relevance between the text content of the first barrage and the video content of the target video clip; and value information of the first barrage, wherein the value information is determined based on the word frequency of at least one word included in the first barrage in the multiple first barrages.

[0099] According to some embodiments, the first correlation between the text content of each first barrage and the video content of the target video clip is obtained using a first determination module, and the first determination module is configured to: input the text content of the first barrage and one or more video frames of the target video clip into a correlation evaluation model to obtain the first correlation output by the correlation evaluation model, wherein the correlation evaluation model is trained using sample text content and sample video clips annotated with a reference first correlation.

[0100] According to some embodiments, the relevance evaluation model includes a text encoding network and a video content encoding network, and the first determination module includes: a first input submodule, configured to input the text content of the first barrage into the text encoding network; and a second input submodule, configured to input one or more video frames of the target video segment into the video content encoding network, so that the relevance evaluation model can determine the first relevance based on the encoding results of the text encoding network and the encoding results of the video content encoding network.

[0101] According to some embodiments, the first determination unit 402 is further configured to: in response to determining that the multiple first barrages include user barrages sent by a target user, for each first barrage in the multiple first barrages except the user barrage, determine a first score for the first barrage based on a second correlation between the text content of the user barrage and the text content of the first barrage, and wherein the display unit 406 is configured to: display the target number of target barrages when playing the target video clip to the target user.

[0102] According to some embodiments, the first determination unit 402 is configured to: input the text content of the user barrage and the text content of the first barrage into a first language model to obtain the second relevance output by the first language model, wherein the first language model is obtained by fine-tuning a large language model using sample text content annotated with a reference second relevance.

[0103] According to some embodiments, the second determination unit 403 is configured to: input the target video clip into a video evaluation model to obtain the second score output by the video evaluation model, wherein the video evaluation model is trained using sample video clips annotated with a reference second score.

[0104] According to some embodiments, the acquisition unit 401 includes: an acquisition subunit, configured to acquire multiple initial barrages for the target video clip; and a determination subunit, configured to determine the multiple first barrages from the multiple initial barrages, wherein the semantic similarity between any two first barrages among the multiple first barrages is not higher than a similarity threshold.

[0105] According to some embodiments, the determination subunit includes: an input module, configured to input the multiple initial barrages into a second language model to obtain multiple initial barrage sets output by the second language model, wherein the semantic similarity between the initial barrages in each initial barrage set is higher than the similarity threshold, and the second language model is obtained by fine-tuning the large language model using multiple sample initial barrages; and a second determination module, configured to determine the multiple first barrages based on the multiple initial barrage sets.

[0106] According to some embodiments, the device 400 also includes: a fifth determination unit, configured to determine, for each target barrage of the target number of target barrages, a corresponding first number of the target barrages, wherein the first number indicates the number of initial barrages among the multiple initial barrages whose semantic similarity with the target barrage is higher than the similarity threshold; and a sixth determination unit, configured to determine, based on the first number, display parameters of the target barrage, wherein the display parameters include at least one of text color, text font size, and text transparency, and wherein the display unit 406 is configured to: when playing the target video clip, display the target barrage based on the display parameters of each target barrage.

[0107] According to some embodiments, the device 400 also includes: a seventh determination unit, configured to determine the user barrage as a target barrage in response to determining that the multiple first barrages include a user barrage sent by a target user, and wherein the display unit 406 is configured to: display the target number of target barrages when playing the target video clip to the target user.

[0108] According to some embodiments, the target video segment is obtained using a third determination module, and the third determination module includes: an analysis submodule, configured to perform shot analysis on the original video to determine a plurality of shots included in the original video; a first determination submodule, configured to determine a plurality of video segments based on the plurality of shots; and a second determination submodule, configured to determine the target video segment from the plurality of video segments.

[0109] It should be understood that Figure 4 The operations of the various units of the video bullet screen display device 400 shown in FIG. Figure 2The steps in the video bullet screen display method 200 described above correspond to each other. Therefore, the operations, features and advantages described above for the method 200 are also applicable to the device 400 and the various units included therein. For the sake of brevity, some operations, features and advantages are not repeated here.

[0110] In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0111] According to another aspect of the present disclosure, an electronic device is also provided, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the above-mentioned video barrage display method.

[0112] According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is also provided, wherein the computer instructions are used to enable the computer to execute the above-mentioned video barrage display method.

[0113] According to another aspect of the present disclosure, a computer program product is also provided, including a computer program, wherein the computer program implements the above-mentioned video barrage display method when executed by a processor.

[0114] refer to Figure 5 , a block diagram of an electronic device 500 that can be used as a server or client of the present disclosure will now be described, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and / or required herein.

[0115] like Figure 5As shown, the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0116] Multiple components in the device 500 are connected to the I / O interface 505, including: an input unit 506, an output unit 507, a storage unit 508, and a communication unit 509. The input unit 506 can be any type of device that can input information to the device 500. The input unit 506 can receive input digital or character information and generate key signal input related to user settings and / or function control of the electronic device, and can include but is not limited to a mouse, a keyboard, a touch screen, a track pad, a track ball, a joystick, a microphone, and / or a remote controller. The output unit 507 can be any type of device that can present information, and can include but is not limited to a display, a speaker, a video / audio output terminal, a vibrator, and / or a printer. The storage unit 508 can include but is not limited to a disk, an optical disk. The communication unit 509 allows the device 500 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks, and can include but is not limited to a modem, a network card, an infrared communication device, a wireless communication transceiver, and / or a chipset, such as a Bluetooth™ device, an 802.11 device, a WiFi device, a WiMax device, a cellular communication device, and / or the like.

[0117] The computing unit 501 may be a variety of general and / or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as a video bullet screen display method. For example, in some embodiments, the video bullet screen display method may be implemented as a computer software program, which is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the video bullet screen display method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the video bullet screen display method in any other appropriate manner (e.g., by means of firmware).

[0118] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0119] The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, enables the functions / operations specified in the flow chart and / or block diagram to be implemented. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

[0120] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be 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, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0121] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0122] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), the Internet, and a blockchain network.

[0123] A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship of client and server is generated by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, a server of a distributed system, or a server combined with a blockchain.

[0124] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0125] Although the embodiments or examples of the present disclosure have been described with reference to the accompanying drawings, it should be understood that the above-mentioned methods, systems and devices are merely exemplary embodiments or examples, and the scope of the present invention is not limited by these embodiments or examples, but only by the claims after authorization and their equivalent scope. Various elements in the embodiments or examples can be omitted or replaced by their equivalent elements. In addition, each step can be performed in an order different from that described in the present disclosure. Further, the various elements in the embodiments or examples can be combined in various ways. It is important that with the evolution of technology, many elements described herein can be replaced by equivalent elements that appear after the present disclosure.

Claims

1. A video bullet screen display method, comprising: Obtain multiple first bullet comments for a target video clip; Determine a first score of each first bullet forum among the plurality of first bullet forums based on text content of the first bullet forum, wherein the first score of each first bullet forum is determined by predicting a user's preference for the first bullet forum; Determining a second score of the target video segment based on the video content of the target video segment, wherein the second score is determined by predicting a user's preference for the video content; determining a target quantity based on the second score; In response to determining that the number of the plurality of first barrages exceeds the target number, determining a target number of target barrages from the plurality of first barrages based on the first score of each of the first barrages; and When playing the target video clip, the target number of target bullet screens is displayed.

2. The method of claim 1, wherein: The determining, based on the text content of each first bullet forum among the plurality of first bullet forums, a first score of the first bullet forum comprises: For each first bullet forum among the plurality of first bullet forums, a first score of the first bullet forum is determined based on at least one of the following items: A first relevance between the text content of the first bullet screen and the video content of the target video clip; and The value information of the first barrage, wherein the value information is determined based on the word frequency of at least one word included in the first barrage in the multiple first barrages.

3. The method of claim 2, wherein: The first relevance between the text content of each first bullet screen and the video content of the target video segment is determined in the following manner: The text content of the first barrage and one or more video frames of the target video clip are input into a relevance evaluation model to obtain the first relevance output by the relevance evaluation model, wherein the relevance evaluation model is trained using sample text content and sample video clips annotated with a reference first relevance.

4. The method of claim 3, wherein: The relevance evaluation model includes a text encoding network and a video content encoding network, and the inputting the text content of the first bullet comment and one or more video frames of the target video segment into the relevance evaluation model to obtain the first relevance output by the relevance evaluation model includes: Inputting the text content of the first bullet screen into the text encoding network; and One or more video frames of the target video segment are input into the video content encoding network, so that the relevance evaluation model can determine the first relevance based on the encoding result of the text encoding network and the encoding result of the video content encoding network.

5. The method according to any one of claims 2 to 4, wherein: The determining, based on the text content of each first bullet forum among the plurality of first bullet forums, a first score of the first bullet forum comprises: In response to determining that the plurality of first bullet comments include a user bullet comment sent by a target user, for each first bullet comment other than the user bullet comment in the plurality of first bullet comments, determining a first score for the first bullet comment based on a second correlation between text content of the user bullet comment and text content of the first bullet comment, And wherein, displaying the target number of target bullet screens when playing the target video clip includes: When playing the target video clip to the target user, the target number of target bullet comments are displayed.

6. The method of claim 5, wherein: The determining, based on the second correlation between the text content of the user bullet screen and the text content of the first bullet screen, a first score of the first bullet screen comprises: The text content of the user barrage and the text content of the first barrage are input into a first language model to obtain the second relevance output by the first language model, wherein the first language model is obtained by fine-tuning a large language model using sample text content annotated with a reference second relevance.

7. The method according to any one of claims 1 to 6, wherein: The determining, based on the video content of the target video segment, a second score of the target video segment comprises: The target video clip is input into a video evaluation model to obtain the second score output by the video evaluation model, wherein the video evaluation model is trained using sample video clips annotated with a reference second score.

8. The method according to any one of claims 1 to 7, wherein: The obtaining of a plurality of first bullet comments for the target video segment comprises: Obtaining multiple initial comments for the target video segment; and The multiple first barrages are determined from the multiple initial barrages, wherein the semantic similarity between any two first barrages in the multiple first barrages is not higher than a similarity threshold.

9. The method of claim 8, wherein: The determining the plurality of first bullet comments from the plurality of initial bullet comments comprises: Inputting the multiple initial bullet comments into a second language model to obtain multiple initial bullet comment sets output by the second language model, wherein the semantic similarity between the initial bullet comments in each initial bullet comment set is higher than the similarity threshold, and the second language model is obtained by fine-tuning a large language model using the multiple sample initial bullet comments; and Based on the multiple initial barrage sets, the multiple first barrages are determined.

10. The method according to claim 8 or 9, further comprising: For each target bullet screen of the target number of target bullet screens, Determine a first number corresponding to the target barrage, wherein the first number indicates the number of initial barrages among the multiple initial barrages whose semantic similarity with the target barrage is higher than the similarity threshold; and Based on the first number, determining display parameters of the target bullet screen, wherein the display parameters include at least one of text color, text font size, and text transparency, And wherein, displaying the target number of target bullet screens when playing the target video clip includes: When playing the target video clip, the target bullet screen is displayed based on the display parameters of each target bullet screen.

11. The method according to any one of claims 1 to 10, further comprising: In response to determining that the plurality of first bullet screens include a user bullet screen sent by a target user, determining the user bullet screen as a target bullet screen, And wherein, displaying the target number of target bullet screens when playing the target video clip includes: When playing the target video clip to the target user, the target number of target bullet comments are displayed.

12. The method according to any one of claims 1 to 11, wherein: The target video segment is obtained in the following manner: Performing shot analysis on an original video to determine a plurality of shots included in the original video; Based on the multiple storyboards, determining multiple video clips; as well as The target video segment is determined from the multiple video segments.

13. A video bullet screen display device, An acquisition unit, configured to acquire a plurality of first bullet comments for a target video segment; The first determining unit is configured to determine a first score of each first bullet forum in the plurality of first bullet forums based on the text content of the first bullet forum, wherein: The first score of each first bullet screen is determined by predicting the user's preference for the first bullet screen; A second determining unit is configured to determine a second score of the target video segment based on the video content of the target video segment, wherein the second score is determined by predicting a user's preference for the video content; a third determining unit, configured to determine a target quantity based on the second score; A fourth determining unit is configured to determine a target number of target barrages from the plurality of first barrages based on the first score of each of the first barrages in response to determining that the number of the plurality of first barrages exceeds the target number; and The display unit is configured to display the target number of target bullet screens when playing the target video clip.

14. The device of claim 13, wherein: The first determining unit is configured to: For each first bullet forum among the plurality of first bullet forums, a first score of the first bullet forum is determined based on at least one of the following items: A first relevance between the text content of the first bullet screen and the video content of the target video clip; and The value information of the first barrage, wherein the value information is determined based on the word frequency of at least one word included in the first barrage in the multiple first barrages.

15. The device of claim 14, wherein: The first relevance between the text content of each first bullet screen and the video content of the target video segment is obtained by using a first determination module, and the first determination module is configured as follows: The text content of the first barrage and one or more video frames of the target video clip are input into a relevance evaluation model to obtain the first relevance output by the relevance evaluation model, wherein the relevance evaluation model is trained using sample text content and sample video clips annotated with a reference first relevance.

16. The device of claim 15, wherein: The relevance evaluation model includes a text encoding network and a video content encoding network, and the first determination module includes: A first input submodule is configured to input the text content of the first bullet screen into the text encoding network; and The second input submodule is configured to input one or more video frames of the target video segment into the video content encoding network, so that the relevance evaluation model can determine the first relevance based on the encoding results of the text encoding network and the encoding results of the video content encoding network.

17. The device according to any one of claims 14 to 16, wherein: The first determining unit is further configured to: In response to determining that the plurality of first bullet comments include a user bullet comment sent by a target user, for each first bullet comment other than the user bullet comment in the plurality of first bullet comments, determining a first score for the first bullet comment based on a second correlation between text content of the user bullet comment and text content of the first bullet comment, And wherein the display unit is configured as: When playing the target video clip to the target user, the target number of target bullet comments are displayed.

18. The device of claim 17, wherein: The first determining unit is configured to: The text content of the user barrage and the text content of the first barrage are input into a first language model to obtain the second relevance output by the first language model, wherein the first language model is obtained by fine-tuning a large language model using sample text content annotated with a reference second relevance.

19. The device according to any one of claims 13 to 18, wherein: The second determining unit is configured to: The target video clip is input into a video evaluation model to obtain the second score output by the video evaluation model, wherein the video evaluation model is trained using sample video clips annotated with a reference second score.

20. The device according to any one of claims 13 to 19, wherein: The acquisition unit comprises: an acquisition subunit, configured to acquire a plurality of initial bullet comments for a target video segment; and The determination subunit is configured to determine the multiple first barrages from the multiple initial barrages, wherein the semantic similarity between any two first barrages in the multiple first barrages is not higher than a similarity threshold.

21. The device of claim 20, wherein: The determining subunit comprises: an input module configured to input the plurality of initial bullet comments into a second language model to obtain a plurality of initial bullet comment sets output by the second language model, wherein the semantic similarity between the initial bullet comments in each initial bullet comment set is higher than the similarity threshold, and the second language model is obtained by fine-tuning a large language model using the plurality of sample initial bullet comments; and The second determination module is configured to determine the multiple first barrages based on the multiple initial barrage sets.

22. The apparatus of claim 20 or 21, further comprising: a fifth determining unit, configured to determine, for each target barrage of the target number of target barrages, a first number corresponding to the target barrage, wherein the first number indicates the number of initial barrages among the multiple initial barrages whose semantic similarity with the target barrage is higher than the similarity threshold; and a sixth determining unit, configured to determine a display parameter of the target bullet comment based on the first number, wherein the display parameter includes at least one of text color, text font size, and text transparency; And wherein the display unit is configured as: When playing the target video clip, the target bullet screen is displayed based on the display parameters of each target bullet screen.

23. The apparatus of any one of claims 13 to 22, further comprising: a seventh determining unit, configured to, in response to determining that the plurality of first bullet screens include a user bullet screen sent by a target user, determine the user bullet screen as a target bullet screen, And wherein the display unit is configured as: When playing the target video clip to the target user, the target number of target bullet comments are displayed.

24. The device according to any one of claims 13 to 23, wherein: The target video segment is obtained by using a third determination module, and the third determination module includes: An analysis submodule, configured to perform shot analysis on the original video to determine a plurality of shots included in the original video; A first determining submodule is configured to determine a plurality of video segments based on the plurality of storyboards; and The second determining submodule is configured to determine the target video segment from the multiple video segments.

25. An electronic device comprising: at least one processor; as well as a memory communicatively coupled to the at least one processor; in The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 12.

26. A non-transitory computer-readable storage medium storing computer instructions, wherein: The computer instructions are used to make a computer execute the method according to any one of claims 1-12.

27. A computer program product comprising a computer program, wherein: The computer program implements the method according to any one of claims 1 to 12 when executed by a processor.

Images