Intelligent advertisement shielding method for cloud game

Abstract

The invention discloses an intelligent advertisement shielding method for a cloud game, which comprises the following steps: after the cloud game is started, a frame collector of a cloud server starts to collect each frame of picture in a game running process, after the frame collector receives the frame picture of the game, an AI large model image-text detection is called, the picture is analyzed, and the frame picture of the game is obtained; whether the frame is an advertisement picture frame or not is judged, if the frame is an advertisement frame, the frame collector carries out frame dropping processing on the frame and carries out frame supplementing operation according to the previous frame of the game, after a frame buffer area in the frame collector is filled, frame data are sent to the encoder to be processed, the encoder carries out encoding compression processing on the received picture frame, and the frame data are sent to the frame collector; according to the scheme, AI is used for intelligently detecting each frame of picture of a game, and if it is judged through AI large model calculation that the frame is a picture with an implanted advertisement, the frame is discarded, the advertisement picture is shielded, and the privacy of a user is protected.

Description

Technical Field

[0001] The present invention relates to the technical field of AI, and in particular to a method for intelligently shielding advertisements in cloud games. Background Art

[0002] Cloud games run on cloud servers, which compress and encode the running screen of the game and push it to the user's terminal device through video stream for display. Image and text recognition uses computer vision and deep learning technology to automatically extract text information from images. Frame interpolation is a video processing technology that increases the frame rate of the video and improves the overall smoothness of the video by inserting new frames between the original video frames.

[0003] The shortcomings of existing cloud games: When users are playing games, various advertisements will be implanted in the game and cannot be actively blocked. Currently, most third-party software blocks advertisements by analyzing the user terminal access IP. All user access IPs will be analyzed by third-party servers. If the third-party server determines that the IP is an advertising server IP, it will intercept it, otherwise it will allow the request.

[0004] There are many disadvantages to using a third-party server for IP interception. First, all user requests will pass through the third-party server, which poses a risk of user privacy leakage. Second, the third-party server often misjudges the IP, affecting the user's actual gaming experience. Summary of the invention

[0005] The purpose of the present invention is to provide a method for intelligently shielding advertisements in cloud games. The present invention integrates cloud games and AI graphics and uses AI technology to detect each frame of the cloud game in real time. Frames determined to be advertisements are discarded, and video frame filling technology is used to fill in the lost frames. Advertising images in the game process can be accurately removed without reducing the overall frame rate of the game.

[0006] To achieve the above object, the present invention provides the following technical solutions: A cloud game intelligent advertising shielding method, comprising the following steps: S1: The user terminal starts the cloud game, the central node distributes the game node information, the cloud server starts the game, and the terminal and the cloud establish a connection; S2: The cloud server collects each frame of the game and puts it into the frame collector buffer; S3: After the frame collector receives the game screen frame, it calls the AI ​​large model image and text detection for each frame; S4: When the frame is determined to be an advertisement picture, it is an advertisement frame and the frame collector is fed back; S5: The frame collector discards the advertisement frame and performs a frame-filling operation based on the previous game frame. S6: When the frame grabber buffer is full, push the batch of frames to the codec; S7: The codec encodes and compresses the received game frame and pushes it to the user terminal; S8: The user terminal receives the video stream, performs decoding operations, and renders the game screen.

[0007] Preferably, the S4 is an advertisement frame, the frame buffer queue performs frame loss processing on the frame, and performs a frame supplement operation with reference to the previous game frame in the frame queue.

[0008] Preferably, if the frame in S4 is a game frame, the frame is normally put into the frame queue and waits to be sent.

[0009] Preferably, when the frame buffer queue in S6 is full, the frame data in the queue is sent to the encoder.

[0010] Preferably, after receiving the frame data, the encoder in S7 encodes the frame data, and after encoding, compresses the frame data into a unified video stream format and sends it to the terminal device.

[0011] Preferably, decoding is performed in S8, the decoded data is rendered, and the game screen is displayed, and the game screen is removed at this time.

[0012] In summary, due to the adoption of the above technology, the beneficial effects of the present invention are: When the present invention runs a game on a cloud server, the running screen of the game will be collected frame by frame by the system to the cloud game frame collector. Whenever the frame collector receives a frame, the AI ​​large model image and text detection method is called to determine the frame as an advertisement frame. If the determination result is an advertisement frame, the frame is discarded, and a frame supplement operation is performed based on the previous game frame. When the frame buffer of the collector is full, the picture frames in the buffer are sent to the cloud game encoder for encoding and compression. After the encoder is compressed, the compressed video stream is pushed to the cloud game terminal device. After receiving the game video stream, the terminal device decodes and plays it. At this time, the game screen displayed by the terminal device is all without the advertisement screen.

[0013] 2. Compared with the existing technology, the main advantage is that the present invention does not need to analyze the user's request IP, which can better protect the user's privacy and security, avoid the risk of game services being blocked by third-party software due to misjudgment of IP, and improve the stability and fluency of the user's game.

[0014] 3. Compared with existing technologies, the main innovation lies in: through real-time image and text analysis of AI large models, it can accurately identify advertising frames during the game process, and adopt frame loss and frame filling technology to remove advertisements from the game without affecting the overall fluency. The integration of AI and cloud gaming technology improves user privacy and the accuracy of ad removal.

[0015] 4. It can analyze various user game screens and image quality issues offline to provide a reference for game model adaptation and problem location. It can intelligently extract relevant game scenes for user complaints. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention, so that other features, purposes and advantages of the present invention become more obvious. The accompanying drawings of the exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings: Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention.

[0018] In the description of the present invention, it is necessary to understand that the terms indicating orientation or positional relationship are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0019] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances of the specification.

[0020] The present invention provides Figure 1 A cloud game intelligent advertising shielding method is shown, comprising the following steps: S1: The user terminal starts the cloud game, the central node distributes the game node information, the cloud server starts the game, and the terminal and the cloud establish a connection; S2: The cloud server collects each frame of the game and puts it into the frame collector buffer; S3: After the frame collector receives the game screen frame, it calls the AI ​​large model image and text detection for each frame; S4: When the frame is determined to be an advertisement picture, it is an advertisement frame and the frame collector is fed back; S5: The frame collector discards the advertisement frame and performs a frame-filling operation based on the previous game frame. S6: When the frame grabber buffer is full, push the batch of frames to the codec; S7: The codec encodes and compresses the received game frame and pushes it to the user terminal; S8: The user terminal receives the video stream, performs decoding operations, and renders the game screen.

[0021] Preferably, the S4 is an advertisement frame, the frame buffer queue performs frame loss processing on the frame, and performs a frame supplement operation with reference to the previous game frame in the frame queue.

[0022] Preferably, if the frame in S4 is a game frame, the frame is normally put into the frame queue and waits to be sent.

[0023] Preferably, when the frame buffer queue in S6 is full, the frame data in the queue is sent to the encoder.

[0024] Preferably, after receiving the frame data, the encoder in S7 encodes the frame data, and after encoding, compresses the frame data into a unified video stream format and sends it to the terminal device.

[0025] Preferably, decoding is performed in S8, the decoded data is rendered, and the game screen is displayed, and the game screen is removed at this time.

[0026] A cloud game intelligent advertisement shielding method. When a game is run on a cloud server, the running screen of the game will be collected frame by frame by the system and collected into a cloud game frame collector. Whenever the frame collector receives a frame, it calls the AI ​​large model image and text detection method to determine whether the frame is an advertisement frame. If the determination result is an advertisement frame, the frame is discarded, and a frame supplement operation is performed based on the previous game frame. When the frame buffer of the collector is full, the picture frames in the buffer are sent to the cloud game encoder for encoding and compression. After compression, the encoder pushes the compressed video stream to the cloud game terminal device. After receiving the game video stream, the terminal device decodes and plays it. At this time, the game screen displayed by the terminal device is all without advertisements. Example

[0027] (1) When a user starts a game on a cloud gaming terminal, the request is sent to the cloud gaming center node.

[0028] (2) After receiving the request from step (1), the game center node performs resource scheduling calculations internally, selects the best game node information, and returns it to the terminal device.

[0029] (3) After receiving the game node information returned in step (2), the user terminal device connects to the game node, the cloud game node starts the game, and a connection is established between the terminal and the cloud.

[0030] (4) Step (3) After the cloud game is started, the game node starts the game screen frame collector program.

[0031] (5) The frame grabber running in step (4) collects each frame of the running game and sends each collected frame to the frame buffer queue.

[0032] (6) After the frame buffer queue in step (5) receives a frame of image, it uses the AI ​​large model image and text detection to determine whether the frame of image is an advertisement image and feedback the real-time detection result.

[0033] (7) Determine the detection result of the picture frame in step (6). If it is an advertisement frame, the frame buffer queue performs frame drop processing on the frame and performs frame filling operation with reference to the previous game frame in the frame queue.

[0034] (8) Determine the detection result of the picture frame in step (6). If it is a game frame, put the frame into the frame queue normally and wait for sending.

[0035] (9) When the frame buffer queue in step (5) is full, the frame data in the queue is sent to the encoder.

[0036] (10) After receiving the frame data, the encoder in step (9) encodes the frame data, compresses it into a unified video stream format, and sends it to the terminal device.

[0037] (11) The terminal device receives the video stream sent in step (10), decodes it, renders the decoded data, and displays the game screen. At this time, the game is without the advertisement screen.

[0038] The beneficial effects of this embodiment are: 1. Through the combination of cloud gaming and AI, the screen frames in the game can be directly analyzed to see whether they are advertisements. This can more accurately remove advertisements from the game without the need for users to implant third-party ad removal software, making user terminals more portable and secure.

[0039] 2. AI large model is used to detect image frame information. In the actual operation process, AI will continue to learn and optimize, and the accuracy and efficiency of detection will continue to improve. Compared with third-party ad removal software, which requires manual and regular maintenance of advertising IP, AI detection is more flexible.

[0040] Application scenarios: 1. Intelligent recognition game process requires harmonious pictures and scenes.

[0041] 2. Intelligent game plug-in recognition: by integrating AI large model analysis, it can detect and judge the content of specific scenes.

[0042] value: 1. It can analyze various user game screens and image quality issues offline, providing a reference for game model adaptation and problem location.

[0043] 2. It can intelligently extract relevant game scenes for the scenes complained by users.

[0044] 3. Improve user experience: Cloud gaming intelligently blocks ads, which can significantly improve the player's gaming experience. The presence of ads will interfere with the player's gaming process and affect the gaming experience. By blocking ads, players can enjoy the game more attentively and improve the gaming experience.

[0045] 4. Save time and traffic: Ads not only interrupt the player's game process, but also consume a lot of traffic. Intelligent ad blocking can save players' time and traffic, especially for those players who only have fragmented time and users with limited traffic. This saving is very important. 12.

[0046] 5. Protect privacy: Many ads are pushed by collecting players’ personal information, which may infringe players’ privacy. Smart ad blocking can protect players’ personal information and allow players to enjoy the game more safely.

[0047] 6. Reduce the risk of misoperation: For elderly users, pop-up ads and interference from bad information may lead to misoperation, or even fraud. Smart ad blocking can reduce this risk and create a safe and secure environment for the elderly.

[0048] 7. Prevent fraud: Many advertisements may contain fraudulent information. The elderly are easily targeted by criminals due to their relatively weak ability to identify fraudulent information. Smart ad blocking can effectively prevent the elderly from being defrauded by accidentally touching advertisements and protect their property safety.

[0049] 8. Improve product competitiveness: In the context of an aging society, the demand for home-based elderly care is increasing. The design of intelligent advertising shielding meets the elderly's demand for a pure use environment, thereby enhancing the market competitiveness of the product.

[0050] 9. Promote health management and emotional companionship: The ad-free design allows cloud gaming smart terminals to focus more on their core functions - health management and emotional companionship, providing a comprehensive and personalized home health ecosystem for the elderly.

[0051] After the cloud game is started, the frame collector of the cloud server begins to collect every frame of the game during operation. When the frame collector receives the frame of the game, it calls the AI ​​large model image detection to analyze the image and determine whether it is an advertising frame. If it is an advertising frame, the frame collector will discard the frame and perform frame interpolation based on the previous frame of the game. When the frame buffer in the frame collector is full, the frame data is sent to the encoder for processing. The encoder encodes and compresses the received picture frame, and after uniformly encapsulating it into a specific format, it is sent to the cloud game terminal device. The terminal device decodes the received video stream and renders the game screen. This solution uses the unique operation mode of cloud games and uses AI to intelligently detect each frame of the game. After the AI ​​large model calculates and determines that there is an advertisement embedded in the picture, the frame is discarded and a new frame is inserted using the interpolation technology. In this way, the advertising screen is shielded and user privacy is protected.

Claims

1. A cloud game intelligent advertising shielding method, characterized by: The following steps are involved: S1: The user terminal starts the cloud game, the central node distributes the game node information, the cloud server starts the game, and the terminal and the cloud establish a connection; S2: The cloud server collects each frame of the game and puts it into the frame collector buffer; S3: After the frame collector receives the game screen frame, it calls the AI ​​large model image and text detection for each frame; S4: When the frame is determined to be an advertisement picture, it is an advertisement frame and the frame collector is fed back; S5: The frame collector discards the advertisement frame and performs a frame-filling operation based on the previous game frame. S6: When the frame grabber buffer is full, push the batch of frames to the codec; S7: The codec encodes and compresses the received game frame and pushes it to the user terminal; S8: The user terminal receives the video stream, performs decoding operations, and renders the game screen.

2. A cloud game intelligent advertising shielding method according to claim 1, characterized in that: The S4 is an advertisement frame, and the frame buffer queue performs frame loss processing on the frame, and performs a frame supplement operation with reference to the previous game frame in the frame queue.

3. A cloud game intelligent advertising shielding method according to claim 1, characterized in that: If the frame in S4 is a game frame, the frame is normally put into the frame queue and waits to be sent.

4. A cloud game intelligent advertising shielding method according to claim 1, characterized in that: When the frame buffer queue in S6 is full, the frame data in the queue is sent to the encoder.

5. A cloud game intelligent advertising shielding method according to claim 1, characterized in that: After receiving the frame data, the encoder in S7 encodes the frame data, compresses it into a unified video stream format, and sends it to the terminal device.

6. A cloud game intelligent advertising shielding method according to claim 1, characterized in that: In the step S8, decoding is performed, the decoded data is rendered, and the game screen is displayed. At this time, the game screen is removed from the advertisement screen.

Images