Network panoramic video judgment method and device based on terminal

Abstract

The invention discloses a network panoramic video judgment method and device based on a terminal, and the method comprises the steps: carrying out the sampling of a played online video, and carrying out the image conversion processing of a collected image, so as to obtain an image pixel standard deviation; and determining an effective frame in the image according to the image pixel standard deviation, and judging whether the online video is a panoramic video or not through the effective frame. According to the method, the panoramic video judgment accuracy can be improved, whether the video content is the panoramic video or not can be automatically and quickly judged, and the waiting time of a user is shortened.

Description

Technical Field

[0001] The present invention relates to the technical field of panoramic video recognition, and in particular to a terminal-based network panoramic video judgment method and device. Background Art

[0002] With the rapid development of virtual reality (VR) technology, VR panoramic videos have become an important medium for immersive content, widely used in games, movies, education and training, telemedicine, and other fields. Unlike traditional videos, VR panoramic videos provide a 360-degree perspective, giving users an immersive experience. Currently, various players that play VR panoramic videos require user interaction to switch rendering modes to distinguish VR panoramic videos from traditional videos. They are unable to automatically identify video content and automatically switch rendering modes based on the video content.

[0003] Currently, there are numerous AI image recognition methods. While it's possible to train models capable of identifying panoramic image features using large amounts of panoramic video images, the similarities between panoramic video and traditional 2D video occur during the initial playback phase. Feeding these similar images into a panoramic video recognition model results in low recognition accuracy and high error rates, making it impossible to determine whether a video is panoramic at the start of playback. This can lead to a mismatch between the rendering mode and the video's features, impacting the user's viewing experience.

[0004] The panoramic video recognition model is deployed on the terminal. During actual user use, the panoramic video recognition model needs to determine whether the video is panoramic based on the video content selected by the user. Because the image frame features input to the panoramic video recognition model are not obvious, the recognition accuracy is low.

[0005] Therefore, how to automatically and efficiently judge panoramic videos, reduce the user's waiting time for judgment, and improve the user's viewing experience of panoramic videos is a technical problem that needs to be solved urgently. Summary of the Invention

[0006] The main purpose of the present invention is to provide a terminal-based network panoramic video judgment method and device, which can improve the accuracy of panoramic video judgment, and at the same time automatically and quickly judge whether the video content is a panoramic video, thereby reducing user waiting time.

[0007] In a first aspect, the present application provides a terminal-based network panoramic video judgment method, wherein the method comprises the steps of:

[0008] Based on sampling the played online video, performing image transformation processing on the collected image to obtain the image pixel standard deviation;

[0009] According to the image pixel standard deviation, a valid frame in the image is determined, and whether the online video is a panoramic video is determined by the valid frame.

[0010] In combination with the first aspect above, as an optional implementation method, a video image collector is used to sample the video selected by the user at set intervals;

[0011] Scaling the image of a captured video frame to adjust it to a set pixel;

[0012] The adjusted image is subjected to color space conversion to obtain a grayscale image, and the standard deviation of pixels of the grayscale image is calculated.

[0013] In combination with the first aspect above, as an optional implementation, according to the formula: Calculate the standard deviation of the grayscale image pixels, where x i is the i-th grayscale information, μ is the set pixel average, and N is the pixel size.

[0014] In combination with the first aspect above, as an optional implementation method, a continuous video stream obtained from the played online video is rendered to render the video stream into a plane graphic layer and a spherical graphic layer;

[0015] Based on the result of determining whether it is a panoramic video, the rendered graphics layer is finally displayed.

[0016] In combination with the first aspect above, as an optional implementation, when the user chooses to play a video, a continuous video stream is obtained;

[0017] Establishing a first texture rendering model to render the video stream on a 2D plane to display a first texture graphics layer by default;

[0018] A second texture rendering model is established, and the video stream is rendered on a sphere to hide the second texture graphic layer.

[0019] In combination with the first aspect above, as an optional implementation, if it is determined to be a non-panoramic video, the 2D plane graphics layer is displayed, the spherical graphics layer is closed, and the second texture rendering is stopped;

[0020] If it is determined to be a panoramic video, the spherical graphics layer is displayed, the 2D plane graphics layer is closed, and the first texture rendering is stopped.

[0021] In combination with the first aspect above, as an optional implementation, if the calculated standard deviation is greater than a set threshold, the acquired image is determined to be a valid frame;

[0022] If the calculated standard deviation is less than the set threshold, the acquired image is determined to be an invalid frame.

[0023] In combination with the first aspect above, as an optional implementation, a grayscale image is obtained after scaling and color space conversion of the image, and the number of unique pixels in the image is calculated using the grayscale image;

[0024] If the number of unique pixels is greater than a set standard value, the acquired image is determined to be a valid frame, otherwise it is an invalid frame.

[0025] In combination with the first aspect above, as an optional implementation method, based on a trained panoramic video recognition model;

[0026] The recognized valid frames are input into the panoramic video recognition model, so as to use the panoramic video recognition model to determine whether the online video is a panoramic video.

[0027] In a second aspect, the present application provides a terminal-based network panoramic video judgment device, which includes:

[0028] A processing module, configured to perform image transformation processing on the collected image based on sampling the played online video to obtain the image pixel standard deviation;

[0029] A discrimination module is used to determine a valid frame in the image according to the standard deviation of the image pixels, and to judge whether the online video is a panoramic video through the valid frame.

[0030] This application provides a terminal-based method and device for determining panoramic network videos. The method includes the following steps: sampling a played online video to perform image transformation processing on the captured image to obtain the image pixel standard deviation; determining valid frames in the image based on the image pixel standard deviation; and determining whether the online video is a panoramic video based on the valid frames. This application can improve the accuracy of panoramic video identification and automatically and quickly determine whether video content is panoramic, reducing user waiting time.

[0031] It should be understood that the foregoing general description and the following detailed description are exemplary only and are not restrictive of the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0033] Figure 1 This is a flow chart of a terminal-based network panoramic video judgment method provided in an embodiment of the present application;

[0034] Figure 2 Schematic diagram of a terminal-based network panoramic video judgment device provided in an embodiment of the present application;

[0035] Figure 3 This is a flowchart of the panoramic video rendering and display provided in an embodiment of the present application. DETAILED DESCRIPTION

[0036] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. In the following description, when referring to the drawings, like numbers in different figures represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all possible embodiments consistent with the present invention. Rather, they are merely examples of apparatus and methods consistent with certain aspects of the present invention, as detailed in the appended claims.

[0037] Furthermore, the drawings are merely schematic illustrations of the present disclosure and are not necessarily drawn to scale. Some of the blocks shown in the drawings are functional entities that do not necessarily correspond to physically or logically separate entities.

[0038] The embodiments of the present application are further described in detail below with reference to the accompanying drawings.

[0039] Reference Figure 1 , Figure 1 FIG. 1 is a flow chart of a terminal-based network panoramic video judgment method provided by the present invention. Figure 1 As shown, the method includes the steps of:

[0040] Step S101: Based on sampling the played online video, image transformation processing is performed on the collected image to obtain the image pixel standard deviation.

[0041] Specifically, a video image collector is used to sample the video selected by the user at set intervals;

[0042] Scaling the image of a captured video frame to adjust it to a set pixel;

[0043] The adjusted image is subjected to color space conversion to obtain a grayscale image, and the standard deviation of pixels of the grayscale image is calculated.

[0044] According to the formula: Calculate the standard deviation of the grayscale image pixels, where x i is the i-th grayscale information, μ is the set pixel average, and N is the pixel size.

[0045] For easier understanding, let's take an example. A video image collector is started. Based on the video URL selected by the user, it requests a small segment of the video stream at regular intervals, forward or backward. A complete frame of image is sufficient. The captured image is resized to 128*128 pixels. The image is sampled at intervals within the user-selected video. The captured image is then scaled, converted to a different color space, grayscale deviation calculated, and unique pixel determination is performed to determine if it is a valid frame.

[0046] In one embodiment, after a video frame is captured, it is first scaled to 128*128 pixels. A three-dimensional array, frame, is used to represent a color frame, including the RGB features, number of rows, and number of columns for each pixel. The calculation process for converting the three-dimensional array, frame, to a grayscale image is: Y = 0.299*R + 0.587*G + 0.114*B. Here, R, G, and B are the RGB features of the pixels in the three-dimensional array, and the result Y is the grayscale information. The grayscale information Y of each pixel is composed of the resulting 128*128 array, denoted as gray_frame.

[0047] Calculate the average value of the 128*128 array gray_frame.

[0048]

[0049] where x i is the i-th grayscale information, N is 128*128=16384. The average value is recorded as μ.

[0050] According to the formula: Calculate the standard deviation of the grayscale image pixels.

[0051] Step S102: determining valid frames in the image based on the standard deviation of the image pixels, and judging whether the online video is a panoramic video based on the valid frames.

[0052] Specifically, if the calculated standard deviation is greater than a set threshold, the acquired image is determined to be a valid frame;

[0053] If the calculated standard deviation is less than the set threshold, the acquired image is determined to be an invalid frame.

[0054] In one embodiment, a grayscale image is obtained after scaling and color space conversion of the image, and the number of unique pixels in the image is calculated using the grayscale image;

[0055] If the number of unique pixels is greater than the set standard value, the acquired image is determined to be a valid frame, otherwise it is an invalid frame (it needs to be explained that because the unique pixels in the color image, including color information, have relatively large pixel differences, the number of unique pixels is relatively small, and it is not easy to determine whether it is a valid frame. Dimensionality reduction to a grayscale image not only retains the original color pixel differences, but also reduces the dimensional information of the compared unique pixels. The number of unique pixels obtained will increase, making it easier to determine whether it is a valid frame.).

[0056] Panoramic video recognition model based on training;

[0057] The recognized valid frames are input into the panoramic video recognition model, so as to use the panoramic video recognition model to determine whether the online video is a panoramic video.

[0058] In one embodiment, the generated 128*128 grayscale information array, gray_frame, is downsampled to a 32*32 array, denoted as small_gray_frame. Each element in small_gray_frame is calculated to represent a pixel value. This yields a one-dimensional array containing all unique pixel values ​​in small_gray_frame, sorted in ascending order. The length of this one-dimensional array is calculated to obtain the number of unique pixel values, denoted as unique_pixels.

[0059] If the standard deviation σ is greater than 50 (50 is an empirical value) or unique_pixels is greater than the image size 128*128 divided by 16 (16 is an empirical value), then the frame is considered a valid frame. Otherwise, it is an invalid frame. (That is, the image pixel value distribution is discrete, and the RGB is converted to grayscale, resized, and downsampled to determine whether the threshold of the valid frame is reached, and then whether it is a valid frame is determined.)

[0060] In one embodiment, after determining whether the online video is a panoramic video using the valid frames, the method further includes: rendering the continuous video stream obtained from the played online video to render the video stream into a plane graphic layer and a spherical graphic layer; and determining the final rendered graphic layer to be displayed based on the determination result of whether the video is a panoramic video. Specifically, when the user selects to play the video, the continuous video stream is obtained;

[0061] Establishing a first texture rendering model to render the video stream on a 2D plane to display a first texture graphics layer by default;

[0062] A second texture rendering model is established, and the video stream is rendered on a sphere to hide the second texture graphic layer.

[0063] If it is determined to be a non-panoramic video, the 2D plane graphics layer is displayed, the spherical graphics layer is closed, and the second texture rendering is stopped;

[0064] If it is determined to be a panoramic video, the spherical graphics layer is displayed, the 2D plane graphics layer is closed, and the first texture rendering is stopped.

[0065] For easier understanding, let's take an example. A continuous video stream is acquired and two texture rendering models are established. The first texture rendering model renders the video image as a 2D plane and displays the 2D plane layer. The second texture rendering model renders the video image as a sphere and hides the sphere layer. The system then waits for the panoramic video recognition model to recognize the image. Based on the recognition result, the system determines which texture rendering model's corresponding graphics layer to display and disables the other texture rendering model's corresponding graphics layer.

[0066] One embodiment also includes training a panoramic video recognition model, which includes designing a panoramic video discrimination algorithm and performing edge computing training on the cloud. The model training process specifically includes: video segmentation in the content distribution system, labeling panoramic videos as panoramic videos, labeling non-panoramic videos as non-panoramic videos, sampling the videos, resizing the sampled images, and generating the panoramic video recognition model using a three-layer convolutional neural network with pooling, a fully connected layer, and binary classification.

[0067] To sum up, when the user starts to choose to play an online video, the valid frames in the video content are quickly obtained and input into the panoramic video recognition model to determine whether the video is a panoramic video, and then adjust the rendering model. This not only realizes automatic and efficient judgment of panoramic videos, but also reduces the user's waiting time for judgment, thereby improving the user's panoramic video viewing experience.

[0068] Reference Figure 2 , Figure 2 FIG. 1 is a schematic diagram of a terminal-based network panoramic video judgment device provided by the present invention, as shown in FIG. Figure 2 As shown, the device includes:

[0069] The processing module 201 is used to perform image transformation processing on the collected images based on sampling of the played online video to obtain the image pixel standard deviation.

[0070] The determination module 202 is configured to determine a valid frame in the image according to the standard deviation of the image pixels, and determine whether the online video is a panoramic video based on the valid frame.

[0071] Furthermore, in a possible implementation, the processing module is further configured to utilize the video image collector to sample the video selected by the user at set intervals;

[0072] Scaling the image of a captured video frame to adjust it to a set pixel;

[0073] The adjusted image is subjected to color space conversion to obtain a grayscale image, and the standard deviation of pixels of the grayscale image is calculated.

[0074] Furthermore, in a possible implementation manner, the processing module is further configured to calculate the output signal according to the formula: Calculate the standard deviation of the grayscale image pixels, where x i is the i-th grayscale information, μ is the set pixel average, and N is the pixel size.

[0075] Furthermore, in a possible implementation, the processing module is further configured to render a continuous video stream obtained from the played online video, so as to render the video stream into a plane graphic layer and a spherical graphic layer;

[0076] Based on the result of determining whether it is a panoramic video, the rendered graphics layer is finally displayed.

[0077] Furthermore, in a possible implementation, the processing module is further configured to obtain a continuous video stream when the user selects to play the video;

[0078] Establishing a first texture rendering model to render the video stream on a 2D plane to display a first texture graphics layer by default;

[0079] A second texture rendering model is established, and the video stream is rendered on a sphere to hide the second texture graphic layer.

[0080] Furthermore, in a possible implementation, the determination module is configured to display the 2D plane graphics layer, close the spherical graphics layer, and stop the second texture rendering if it is determined that the video is not a panoramic video;

[0081] If it is determined to be a panoramic video, the spherical graphics layer is displayed, the 2D plane graphics layer is closed, and the first texture rendering is stopped.

[0082] Furthermore, in a possible implementation manner, the discrimination module is further configured to determine that the acquired image is a valid frame if the calculated standard deviation is greater than a set threshold;

[0083] If the calculated standard deviation is less than the set threshold, the acquired image is determined to be an invalid frame.

[0084] Furthermore, in a possible implementation, the discrimination module is further configured to calculate the number of unique pixels in the image using a grayscale image obtained after scaling and color space conversion processing of the image;

[0085] If the number of unique pixels is greater than a set standard value, the acquired image is determined to be a valid frame, otherwise it is an invalid frame.

[0086] Furthermore, in a possible implementation, the discrimination module is also used based on a trained panoramic video recognition model;

[0087] The recognized valid frames are input into the panoramic video recognition model, so as to use the panoramic video recognition model to determine whether the online video is a panoramic video.

[0088] Reference Figure 3 , Figure 3 The flowchart of the panoramic video rendering display provided by the present invention is shown as follows: Figure 3 As shown:

[0089] S301: When the user chooses to play a video, the playback rendering subsystem and the discrimination subsystem work simultaneously, and then proceed to step S401 and step S501.

[0090] S401: Start the player and obtain a continuous video stream.

[0091] S402: Create a first texture and render the video stream on a 2D plane.

[0092] S403: Displaying the first texture graphic layer.

[0093] S404: Create a second texture and render the video stream on the sphere.

[0094] S405: Hide the second texture graphic layer.

[0095] S406: Wait for the panoramic video judgment result, and proceed to step S407 according to the result.

[0096] S407: Based on the result, if it is a non-panoramic video, display the 2D plane graphics layer, close the spherical graphics layer, and stop the second texture rendering. Similarly, if it is a panoramic video displaying the spherical graphics layer, close the 2D plane graphics layer, and stop the first texture rendering.

[0097] S501: Start the video image collector.

[0098] S502: Sampling the video selected by the user at intervals (eg, 5 seconds).

[0099] S503: Adjust the collected image to 128*128 pixels

[0100] S504: Convert the image from the RGB color space to a grayscale image.

[0101] S505: Calculate the standard deviation of the grayscale image.

[0102] S506: Calculate the number of unique pixels.

[0103] S507: Determine whether it is a valid frame based on the standard deviation or the number of unique pixels. If yes, proceed to step S508; if not, proceed to step S502.

[0104] S508: Input the valid frame as a content source to the trained discriminant model.

[0105] S509: Determine whether it is a panoramic video and notify S406 of the result.

[0106] In summary, the beneficial effects of this application include:

[0107] 1. The AI ​​algorithm model trained on non-terminals is combined with the calculation and judgment of valid frames on the terminal to improve the accuracy of panoramic video recognition.

[0108] 2. Through the joint solution of the playback rendering subsystem and the judgment subsystem, it is possible to automatically and quickly determine whether the video content is a panoramic video, reducing user waiting time.

[0109] Based on the same inventive concept, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, all or part of the method steps of the above method are implemented.

[0110] The present application implements all or part of the processes in the above-mentioned method, and can also be completed by instructing related hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, it can implement the steps of the above-mentioned various method embodiments. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device that can carry computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0111] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program running on the processor, and when the processor executes the computer program, all or part of the method steps in the above method are implemented.

[0112] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor. The processor is the control center of a computer device and connects various parts of the entire computer device using various interfaces and lines.

[0113] The memory can be used to store computer programs and / or modules. The processor implements various functions of the computer device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); in addition, the memory can include a high-speed random access memory and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0114] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, servers, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage and optical storage, etc.) containing computer-usable program code.

[0115] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), servers and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0116] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0117] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0118] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

Claims

1. A terminal-based network panoramic video judgment method, characterized in that: include: Based on sampling the played online video, performing image transformation processing on the collected image to obtain the image pixel standard deviation; A valid frame in the image is determined according to the image pixel standard deviation, and whether the online video is a panoramic video is determined by the valid frame.

2. The method according to claim 1, characterized in that The method of performing image transformation processing on the collected image based on sampling the played online video to obtain the image pixel standard deviation includes: The video image collector is used to sample the video selected by the user at set intervals; Scaling the image of a captured video frame to adjust it to a set pixel; The adjusted image is subjected to color space conversion to obtain a grayscale image, and the standard deviation of pixels of the grayscale image is calculated.

3. The method according to claim 2, characterized in that include: According to the formula: Calculate the standard deviation of the grayscale image pixels, where x i is the i-th grayscale information, μ is the set pixel average, and N is the pixel size.

4. The method according to claim 1, wherein After determining whether the online video is a panoramic video through the valid frame, the method includes: Rendering a continuous video stream obtained from a played online video to render the video stream into a plane graphic layer and a spherical graphic layer; Based on the result of determining whether it is a panoramic video, the rendered graphics layer is finally displayed.

5. The method according to claim 4, characterized in that The step of rendering the continuous video stream obtained from the played online video to render the video stream into a plane graphic layer and a spherical graphic layer includes: When the user chooses to play a video, obtain a continuous video stream; Establishing a first texture rendering model to render the video stream on a 2D plane to display a first texture graphics layer by default; A second texture rendering model is established, and the video stream is rendered on a sphere to hide the second texture graphic layer.

6. The method according to claim 4, characterized in that The process of determining the final display rendering graphics layer based on the judgment result of whether the video is a panoramic video includes: If it is determined to be a non-panoramic video, the 2D plane graphics layer is displayed, the spherical graphics layer is closed, and the second texture rendering is stopped; If it is determined to be a panoramic video, the spherical graphics layer is displayed, the 2D plane graphics layer is closed, and the first texture rendering is stopped.

7. The method according to claim 1, characterized in that Determining a valid frame in the online video according to the image pixel standard deviation includes: If the calculated standard deviation is greater than the set threshold, the acquired image is determined to be a valid frame; If the calculated standard deviation is less than the set threshold, the acquired image is determined to be an invalid frame.

8. The method according to claim 1, characterized in that The determining of the valid frames in the online video further includes: A grayscale image obtained after scaling and color space conversion of the image is performed, and the number of unique pixels in the image is calculated using the grayscale image; If the number of unique pixels is greater than a set standard value, the acquired image is determined to be a valid frame, otherwise it is an invalid frame.

9. The method according to claim 1, characterized in that The determining whether the online video is a panoramic video based on the valid frame includes: Panoramic video recognition model based on training; The recognized valid frames are input into the panoramic video recognition model, so as to use the panoramic video recognition model to determine whether the online video is a panoramic video.

10. A terminal-based network panoramic video judgment device for implementing the terminal-based network panoramic video judgment method according to any one of claims 1 to 9, characterized in that: include: A processing module, configured to perform image transformation processing on the collected image based on sampling the played online video to obtain the image pixel standard deviation; A discrimination module is used to determine a valid frame in the image according to the standard deviation of the image pixels, and to judge whether the online video is a panoramic video through the valid frame.

Images