Video director method and device, medium and program product

By obtaining user operation instructions in PGC production, determining and tracking target objects, evaluating and displaying video images with the best viewing angle, the problem of poor audience autonomy in the director switching method is solved, and the user experience is improved.

CN119011885APending Publication Date: 2024-11-22MIGU VIDEO TECH CO LTD +2
View PDF 5 Cites 0 Cited by

Patent Information

Application Number
CN202410924881.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-22

AI Technical Summary

Technical Problem

In PGC production, the audience has poor autonomy in the director switching method, low user experience, and cannot independently choose the perspective of watching the game.

Method used

By obtaining the user's operation instructions on the live video screen, determining the target area and target object, detecting and tracking the object in the video screen of multiple perspectives, evaluating and displaying the video screen of the optimal perspective, providing the most important perspective according to the user's preferences Good perspective.

Benefits of technology

It improves the user's independent selection ability and user experience, and can display the best perspective of the area of ​​interest according to the user's preferences.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119011885A_ABST
    Figure CN119011885A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of artificial intelligence, and provides a video director method and device, a medium and a program product, and the method comprises the steps: obtaining an operation instruction for a currently displayed video image of a live broadcast scene; determining a target area in the video picture according to the operation instruction, and obtaining a target object in the target area; detecting and tracking a target object in all video pictures of different visual angles of the live scene; and displaying the video picture containing the target object. Therefore, the optimal view angle of the interested area can be displayed to the user according to the preference of the user, and the use experience of the user is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The embodiments of the present application relate to the field of artificial intelligence technology, and in particular to a video directing method, device, medium and program product. Background Art

[0002] At present, in the production of PGC (Professionally Generated Content), the director switching method is usually controlled by the director to select and control the existing multi-channel front-end (live video) signals. The switching time and the selection of video channels are based on the director's subjective choice. The audience can only watch the viewing angle set by the front end (such as the end where the director team is located) on the playback terminal, and cannot independently choose the viewing angle hotspot they want to watch. Summary of the invention

[0003] The embodiments of the present application provide a video directing method, device, medium and program product to solve the technical problems of poor viewer autonomy and low user experience in the directing switching method in the related art.

[0004] To solve the above technical problems, the embodiments of the present application provide the following aspects:

[0005] In a first aspect, an embodiment of the present application provides a video directing method, the method comprising:

[0006] Obtaining operation instructions for the currently displayed video screen of the live broadcast scene;

[0007] Determine a target area in the video image according to the operation instruction, and obtain a target object in the target area;

[0008] Detecting and tracking the target object in all video images of different viewing angles of the live scene;

[0009] A video screen containing the target object is displayed.

[0010] Optionally, acquiring the target object in the target area includes:

[0011] Extracting target image features in the target area;

[0012] The target object corresponding to the target image feature is determined according to the target image feature.

[0013] Optionally, acquiring the target object in the target area includes:

[0014] Extracting target image features in the target area;

[0015] In a preset label library, the target object corresponding to the target image feature is searched according to the target image feature, wherein the label library stores a one-to-one correspondence between image features and objects.

[0016] Optionally, detecting and tracking the target object in all video images of different perspectives of the live broadcast scene includes:

[0017] Determining the position and motion vector of the target object;

[0018] The target object is tracked based on the position and the motion vector.

[0019] Optionally, displaying a video screen containing the target object includes:

[0020] Perform at least one of the following evaluations on the video picture containing the target object: content relevance evaluation, video clarity evaluation, video stability evaluation, historical switching mode evaluation, and target switching curve evaluation, and obtain a total evaluation result;

[0021] Determine an optimal overall evaluation result according to the overall evaluation result, and determine a video picture containing the target object corresponding to the optimal overall evaluation result as a target video picture;

[0022] The target video picture is displayed.

[0023] Optionally, determining the optimal overall evaluation result according to the overall evaluation result includes:

[0024] Determine the total evaluation result with the highest score among the total evaluation results as the optimal total evaluation result, wherein the number of the total evaluation results with the highest score is one;

[0025] Alternatively, among the total evaluation results, any total evaluation result with the highest score is randomly determined as the optimal total evaluation result, wherein there are multiple total evaluation results with the highest scores.

[0026] Optionally, displaying a video screen containing the target object includes:

[0027] Get the user's perspective switching instruction;

[0028] According to the view switching instruction, a target video picture is selected from the video pictures containing the target object;

[0029] The target video picture is displayed.

[0030] In a second aspect, an embodiment of the present application provides an electronic device, comprising: a processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the video directing method described in the first aspect.

[0031] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of a video directing method as described in the first aspect are implemented.

[0032] In a fourth aspect, an embodiment of the present application provides a computer program product, including computer instructions, which, when executed by a processor, implement the steps of a video directing method as described in the first aspect.

[0033] Therefore, the method shown in the embodiment of the present application can obtain the operation instruction for the currently displayed video screen of the live scene, determine the target area in the video screen according to the operation instruction, obtain the target object in the target area, detect and track the target object in the video screens of all different perspectives of the live scene, and display the video screen containing the target object. The technical problem of poor audience autonomy and low user experience in the director switching method in the related art is solved, and the best perspective of the area of ​​interest to the user can be shown to the user according to the user's preference, thereby improving the user's experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present application. Also, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0035] Figure 1 A flowchart of a video directing method provided in an embodiment of the present application;

[0036] Figure 2 A schematic diagram of obtaining a zoom or click action of a viewer on a video played on a screen during a viewer watching a live broadcast provided in an embodiment of the present application;

[0037] Figure 3 A schematic diagram of a method of performing a secondary confirmation using a dotted frame or highlighting in a telescoping action or a clicking action provided in an embodiment of the present application;

[0038] Figure 4 A flowchart of a video directing method provided in an embodiment of the present application;

[0039] Figure 5A structural block diagram of a video directing device provided in an embodiment of the present application;

[0040] Figure 6 A structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0042] Figure 1 A video directing method according to an embodiment of the present application is shown. Figure 1 As shown, the method includes:

[0043] Step S101: obtaining an operation instruction for a currently displayed video screen of a live broadcast scene;

[0044] Step S102: determining a target area in the video image according to the operation instruction, and acquiring a target object in the target area;

[0045] Step S103: Detect and track the target object in all video images of different viewing angles of the live scene;

[0046] Step S104: display the video image containing the target object.

[0047] In step S101, the hotspot area of ​​interest of the user who is currently watching the live broadcast can be determined by obtaining the operation instructions of the user for the currently displayed screen. Specifically, the terminal player can upload information such as: the touch position of the hand on the screen, the touch duration of the hand on the screen, the touch duration can be determined by the UTC (Coordinated Universal Time) time difference between the start and end of the touch, the touch mode of the hand on the screen (click or stretch, that is, the audience's stretching or clicking action on the video played on the screen is obtained during the audience's viewing of the live broadcast, such as Figure 2 As shown in the figure, the hot spots of interest can be obtained by using the zoom action coverage area to determine the viewer's interest point. The zoom action coverage area refers to the area displayed on the screen after the viewer stretches the video being played and places the existing interest point area as close to the center of the screen as possible. The video content in the area where the viewer clicks on the screen can be determined as the viewer's interest point. In addition, the zoom action and click action can be confirmed twice with a dotted box or highlight, such as Figure 3 shown.

[0048] It should be noted that each zooming or clicking action of the audience can generate a new hot video clip. The generation method is: after confirming the start and end time of the zoom or click, the video is edited in the corresponding time period, and the editing can be based on the start time of GOP (Group of Pictures) and the end time of GOP as the boundary. It should also be noted that this editing method is not limited to videos with a fixed perspective, but can be applied to videos of different perspectives within the same UTC time period. If the user switches to a new perspective and determines the area of ​​interest in the current video display interface corresponding to the perspective, the area of ​​interest can be edited.

[0049] In step S102, optionally, obtaining the target object in the target area includes: extracting the target image features in the target area; and determining the target object corresponding to the target image features according to the target image features. Exemplarily, the target image features can be input into a neural network model to obtain the target object corresponding to the target image features. Alternatively, obtaining the target object in the target area includes: extracting the target image features in the target area; and searching for the target object corresponding to the target image features in a preset tag library according to the target image features, wherein the tag library stores a one-to-one correspondence between image features and objects.

[0050] It should be noted that after obtaining the hot video, the video recognition in the video content can be completed through the object identifier. The relevant recognition needs to occur between multiple video live signals, that is, the object identifier needs to perform object recognition on multiple videos. The object identifier is based on deep learning algorithms, such as CNN (Convolutional Neural Network), or RNN (Recurrent Neural Network), or traditional computer vision algorithms, such as feature-based methods. Commonly used equipment for object recognition is a high-performance computer.

[0051] In step S102, if it is necessary to obtain the target object through the neural network model, the neural network model needs to be trained first. Specifically, the features identified in the relevant video can be labeled (manually manually labeled or automatically labeled) and the relevant tags can be entered into the tag library. The tag can be a specific person, object or scene (such as a well-known football player, cat, beach), or an abstract concept (such as happiness, surprise). Taking a well-known football player as an example, the corresponding relationship between the feature and the tag (target object) is: the visual information extracted from the target area, such as jersey color, number and facial features, etc., the corresponding tag is the football player. And multiple frame-level tags can form a video-level tag, that is, form a description of the entire video. The purpose of the tag library is to provide a standardized classification system so that the video recognition algorithm has a clear reference when performing tag prediction (i.e., during the model training process, or in the further improvement of the model use process).

[0052] In step S102, the target object corresponding to the target image feature can also be searched according to the target image feature through a preset tag library, wherein the tag library stores a one-to-one correspondence between image features and objects. Thus, the method of obtaining the target object can be flexibly selected according to actual needs.

[0053] In step S103, the target object can be detected and tracked in the video screens of all different perspectives of the live scene, including: determining the position and motion vector of the target object; tracking the target object based on the position and motion vector. It should be noted that after analyzing the features during the live broadcast, the streaming server can retrieve the tags corresponding to the features in the tag library, determine the target object, and then track the target object, and switch the user's perspective to the corresponding video stream. Perspective switching refers to the switching between multi-channel multi-perspective video live signals completed during video production and shooting. The principle of perspective switching is a switching method generated by intelligently selecting the input streams of multiple cameras at the front end of the shooting. And the perspective switching can adopt an algorithm based on motion analysis, which uses the motion information of the object at different perspectives to switch the perspective. First, the object's position information and motion vector information are extracted from the video streams of multiple perspectives using an object detection and tracking algorithm. Among them, the position information can be used to determine the exact position of the object. The object can be separated from the background through the detection algorithm and its position coordinates in the image / video can be determined. These position information can be used to track the movement of the object in real time, such as tracking various targets in the video. Motion vector information can describe the movement mode and speed of the object in time. By calculating the motion changes of an object between consecutive frames, the object's direction and speed can be inferred. These motion vectors are very important for target tracking and can be used to predict the position of the object in future frames and to track and predict it. In this way, the target object can be detected and tracked in all video images of different perspectives in the live scene.

[0054] In step S104 , a video frame including the target object may be displayed.

[0055] In one possible implementation, Figure 4 As shown, step S104, displaying the video screen containing the target object includes:

[0056] Step S401, performing at least one of the following evaluations on the video picture containing the target object: content relevance evaluation, video clarity evaluation, video stability evaluation, historical switching mode evaluation, target switching curve evaluation, and obtaining a total evaluation result;

[0057] Step S402: determining an optimal overall evaluation result according to the overall evaluation result, and determining a video picture containing a target object corresponding to the optimal overall evaluation result as a target video picture;

[0058] Step S403: display the target video screen.

[0059] It should be noted that due to the movement of objects and the changes in different camera images, the position and proportion of the target object in different images change. The motion analysis algorithm is used to enhance the timeliness of image switching based on the detection of moving objects. It is also necessary to select the best viewing angle for switching according to the position change and speed of the object. For example, when the object moves from the current viewing angle to the field of view of other viewing angles, switch to the best new viewing angle that contains the object.

[0060] Among them, determining the optimal total evaluation result according to the total evaluation results includes: determining the total evaluation result with the highest score in the total evaluation results as the optimal total evaluation result, wherein the number of total evaluation results with the highest score is one; or, randomly determining any total evaluation result with the highest score in the total evaluation results as the optimal total evaluation result, wherein the number of total evaluation results with the highest score is multiple. That is to say, when the number of total evaluation results with the highest score is one, it is directly determined as the optimal total evaluation result. If there are multiple total evaluation results with the highest scores, one of them can be arbitrarily selected as the optimal total evaluation result.

[0061] The evaluation method is explained by taking a specific application scenario as an example. Suppose the relevant evaluation function is: Evaluation function Y = α×A+β×B+γ×C+δ×D+ε×E, where α, β, γ, δ and ε are weighting factors used to adjust the importance of each evaluation indicator. A is the content relevance score, B is the clarity score, C is the stability score, D is the historical switching mode score, and E is the target switching curve score. When the value of the evaluation function Y is greater than 1, switch to the corresponding perspective of the evaluation function Y. If there are multiple evaluation functions Y with values ​​greater than 1, the corresponding perspective of the evaluation function Y with the maximum value is selected as the target perspective.

[0062] Evaluation items may include: Content relevance score: used to measure the relevance of the content presented by the perspective to user needs. Object detection, semantic segmentation, key frame extraction and other technologies can be considered to score according to the frequency and importance of the target object, and the trade-off factor is α. Clarity score: used to measure the image quality of the perspective. Image quality assessment techniques such as image contrast, clarity, detail retention and other indicators can be used to score the image. The trade-off factor is β; Stability score: used to measure the image stability of the perspective. The stability of the image is scored according to the jitter of the image. The trade-off factor is γ; Historical switching mode score: used to consider the past perspective switching mode. The historical switching mode can be analyzed and evaluated based on the past switching records and user feedback. For example, if the past switching mode is frequent, the switching frequency can be reduced by lowering the score. The trade-off factor is δ. Target switching curve score: used to evaluate based on the target switching curve. The evaluation can be performed based on the degree of deviation between the expected curve and the current curve of the target switching curve. In a multi-camera video system, the target switching curve evaluation involves evaluating the smoothness and continuity of the perspective switching action. This score is to ensure that the perspective switching is not only visually logical but also in line with the user's viewing comfort. The trade-off factor is ε.

[0063] The method for confirming the relevant trade-off factors can be as follows: the trade-off factors are empirical values, and all the trade-off factors in the default initial state are 1.

[0064] It should be noted that when the corresponding object retrieved in the video stream disappears (the target object cannot be detected), the automatic switching of the perspective stops, but at this time, it can switch to the perspective selected by the user according to the user's perspective switching instruction. That is to say, in another possible implementation method, displaying the video screen containing the target object includes: obtaining the user's perspective switching instruction; selecting the target video screen from the video screen containing the target object according to the perspective switching instruction; and displaying the target video screen to ensure that the user has a high priority in decision-making.

[0065] In summary, the method shown in the embodiment of the present application can utilize the user's actions such as zooming in or clicking on the current display interface while watching the video to confirm the user hot spot area of ​​the video content, complete feature extraction for the user hot spot area, and retrieve the video content based on the acquired features, and switch the perspective of the retrieved content. This solves the technical problems of poor audience autonomy and low user experience in the director switching method in the related art, and can show the user the best perspective of the area of ​​interest according to the user's preferences, thereby improving the user's experience.

[0066] The overall process of the video directing method shown in the embodiment of the present application is now described.

[0067] In the method shown in the embodiment of the present application, the multi-video stream view switching function based on the target object is mainly used to capture the video of the target object from different angles through multiple cameras or video sources, and realize real-time view switching. The steps are as follows:

[0068] 1: Video acquisition and synchronization: Connect multiple cameras or video sources to the system and ensure UTC time synchronization between video streams. PTP (Precision Time Protocol) can be used for synchronization.

[0069] 2: Calibration and positioning: Calibrate each camera or video source and determine their relative positions and angles. This can be achieved by using calibration plates or feature point technology.

[0070] 3: Object detection and tracking: Use object detection and tracking algorithms to identify and track objects in video streams in real time or offline. This can use deep learning algorithms such as convolutional neural networks or recurrent neural networks, or use traditional computer vision algorithms such as feature-based methods.

[0071] 4: Perspective switching: Switch between different cameras or video sources as needed. This can be achieved based on information such as the position, speed, direction of movement of the target object, as well as user instructions or algorithm decisions.

[0072] In a specific application scenario, when a user is watching a football match and sees his favorite football player in a picture, the user stretches and clicks the picture of the football player on the mobile phone; after the user stretches the picture, the hot spot area feature extraction function is triggered, that is, the feature of the current display video interface in the existing live broadcast is extracted, and the target object is determined based on the feature, and then the target object is identified and tracked; after that, when the football player appears, the picture can be switched according to the actual picture situation to maximize the user's interest in watching the football player. And when the picture is switched, an evaluation can be performed to determine the best picture to switch to.

[0073] The relevant evaluation function is as follows: Y = α × A + β × B + γ × C + δ × D + ε × E, where α, β, γ, δ and ε are weighting factors used to adjust the importance of each evaluation indicator. A is the content relevance score, B is the clarity score, C is the stability score, D is the historical switching mode score, and E is the target switching curve score.

[0074] Assume that there are three perspectives and the evaluation function after evaluation at time T is:

[0075] View 1 evaluation function Y1=0.1α+0.1β+0.1γ+0.1δ+0.1ε;

[0076] View 2 evaluation function Y2 = 0.2α + 0.2β + 0.2γ + 0.2δ + 0.2ε;

[0077] View 3 evaluation function Y3 = 0.3α + 0.3β + 0.3γ + 0.3δ + 0.3ε;

[0078] Assume that the currently configured α, β, γ, δ and ε are set to the default value 1. Then the evaluation function Y3 of view angle 3=1.5>1, so it is necessary to switch to view angle 3 at the current moment.

[0079] Therefore, switching from the existing manual directing method to a feature content retrieval directing method with a perspective switching algorithm can greatly improve the efficiency of directing and expand the application scenarios of the solution, from traditional PGC applications to more PUGC (Professionally User Generated Content) and UGC (User Generated Content) applications.

[0080] Figure 5 A video director device is shown, such as Figure 5 As shown, the video directing device 50 includes:

[0081] The acquisition module 501 is used to acquire an operation instruction for the currently displayed video picture of the live broadcast scene;

[0082] An execution module 502 is used to determine a target area in the video image according to the operation instruction, and obtain a target object in the target area;

[0083] Detect and track the target object in all video frames from different perspectives of the live scene;

[0084] Display the video frame containing the target object.

[0085] In a possible implementation, the execution module 502 is further used to extract target image features in the target area;

[0086] The target image features are input into the neural network model to obtain the target object corresponding to the target image features.

[0087] In a possible implementation, the execution module 502 is further used to extract target image features in the target area;

[0088] In a preset label library, a target object corresponding to the target image feature is searched according to the target image feature, wherein a one-to-one correspondence between the image feature and the object is stored in the label library.

[0089] In a possible implementation, the execution module 502 is further configured to determine the position and motion vector of the target object; and track the target object based on the position and motion vector.

[0090] In a possible implementation, the execution module 502 is further configured to perform at least one of the following evaluations on the video picture containing the target object: content relevance evaluation, video clarity evaluation, video stability evaluation, historical switching mode evaluation, and target switching curve evaluation, and obtain a total evaluation result;

[0091] Determine the optimal overall evaluation result according to the overall evaluation result, and determine the video picture containing the target object corresponding to the optimal overall evaluation result as the target video picture;

[0092] Display the target video screen.

[0093] In one possible implementation, the execution module 502 is also used to determine the total evaluation result with the highest score in the total evaluation results as the optimal total evaluation result, where the number of total evaluation results with the highest score is one; or, randomly determine any total evaluation result with the highest score in the total evaluation results as the optimal total evaluation result, where the number of total evaluation results with the highest score is multiple.

[0094] In a possible implementation, the execution module 502 is further configured to obtain a user's perspective switching instruction; select a target video picture from video pictures containing a target object according to the perspective switching instruction; and display the target video picture.

[0095] In summary, the device shown in the embodiment of the present application can utilize the user's actions such as zooming in or clicking on the current display interface during video viewing to confirm the user hotspot area of ​​the video content, complete feature extraction for the user hotspot area, and retrieve the video content based on the acquired features, and switch the perspective of the retrieved content. This solves the technical problems of poor audience autonomy and low user experience in the director switching method in the related art, and can show the user the best perspective of the area of ​​interest according to the user's preferences, thereby improving the user's experience.

[0096] The present application embodiment also provides an electronic device 60, such as Figure 6 As shown, it includes: a processor 601, a memory 602, and a program stored in the memory 602 and executable on the processor 601. When the program is executed by the processor, the steps of a video directing method as shown in the above embodiment are implemented.

[0097] The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of a video directing method shown in the above method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. The computer-readable storage medium is, for example, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0098] An embodiment of the present application also provides a computer program product, including computer instructions. When the computer instructions are executed by a processor, the steps of a video directing method shown in the above method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, they are not repeated here.

[0099] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0100] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0101] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A video directing method, characterized in that: The method comprises: Obtaining operation instructions for the currently displayed video screen of the live broadcast scene; Determine a target area in the video image according to the operation instruction, and obtain a target object in the target area; Detecting and tracking the target object in all video images of different viewing angles of the live scene; A video screen containing the target object is displayed.

2. The method according to claim 1, characterized in that The acquiring of the target object within the target area comprises: Extracting target image features in the target area; The target object corresponding to the target image feature is determined according to the target image feature.

3. The method according to claim 1, characterized in that The acquiring of the target object within the target area comprises: Extracting target image features in the target area; In a preset label library, the target object corresponding to the target image feature is searched according to the target image feature, wherein the label library stores a one-to-one correspondence between image features and objects.

4. The method according to claim 1, characterized in that: Detecting and tracking the target object in all video images of different viewing angles of the live scene includes: Determining the position and motion vector of the target object; The target object is tracked based on the position and the motion vector.

5. The method according to any one of claims 1 to 4, characterized in that The displaying of the video screen containing the target object comprises: Perform at least one of the following evaluations on the video picture containing the target object: content relevance evaluation, video clarity evaluation, video stability evaluation, historical switching mode evaluation, and target switching curve evaluation, and obtain a total evaluation result; Determine an optimal overall evaluation result according to the overall evaluation result, and determine a video picture containing the target object corresponding to the optimal overall evaluation result as a target video picture; The target video picture is displayed.

6. The method according to claim 5, characterized in that Determining the optimal overall evaluation result according to the overall evaluation result includes: Determine the total evaluation result with the highest score among the total evaluation results as the optimal total evaluation result, wherein the number of the total evaluation results with the highest score is one; Alternatively, among the total evaluation results, any total evaluation result with the highest score is randomly determined as the optimal total evaluation result, wherein there are multiple total evaluation results with the highest scores.

7. The method according to any one of claims 1 to 4, characterized in that The displaying of the video screen containing the target object comprises: Get the user's perspective switching instruction; According to the view switching instruction, a target video picture is selected from the video pictures containing the target object; The target video picture is displayed.

8. An electronic device, characterized in that: include: A processor, a memory, and a program stored in the memory and executable on the processor, wherein the program, when executed by the processor, implements the steps of the video directing method as described in any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the video directing method according to any one of claims 1 to 7 are implemented.

10. A computer program product, characterized in that It includes computer instructions, which, when executed by a processor, implement the steps of the video directing method as described in any one of claims 1 to 7.

Citation Information

Patent Citations

  • Target tracking method and device and portable terminal

    CN110197126A

  • Video playing method and device, electronic equipment and computer storage medium

    CN113301351A

  • Free viewpoint visual angle information recommendation method and related equipment

    CN113473157A

  • Video processing method, playing method, related system and storage medium

    CN117440128A

  • tracking A TARGET IN A NETWORK OF CAMERAS

    FR3039919A1