High-definition video multi-screen synchronous transmission system and method

Abstract

The invention relates to the technical field of video synchronous transmission, in particular to a high-definition video multi-screen synchronous transmission system and method, and the method comprises the steps: determining a system synchronous baseline through a state collection and decision module by taking equipment with the maximum expected decoding delay as a reference, and generating a unified timestamp sequence; the video preprocessing module generates a reference synchronization stream forcibly decoded by all the devices based on the baseline, and generates image quality enhancement data for the high-performance devices; the data distribution module respectively transmits a reference stream, enhanced data and a timestamp through a multi-channel mechanism; the client synchronous rendering module fuses enhanced data by high-performance equipment on the basis of decoding the reference stream, and all the equipment performs synchronous rendering strictly according to a unified timestamp; and the dynamic adjustment module continuously monitors the state of the equipment, and updates a synchronization baseline and a timestamp in real time when the delay change exceeds a threshold, so that the frame-level synchronism and the image quality consistency of multi-screen display are simultaneously ensured in a heterogeneous network environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of video synchronous transmission, and particularly relates to a high-definition video multi-screen synchronous transmission system and method. BACKGROUND

[0002] With the popularity of intelligent terminal devices and the in-depth application of multimedia, the demand for synchronously displaying the same high-definition video content on multiple screens (such as smart phones, tablets, smart televisions and video walls) is increasing, which is common in scenes such as home audio and video, video conference, digital signage and immersive exhibition hall. The prior art usually adopts an adaptive bitstream transmission scheme, that is, a server provides multiple video streams with different bit rates, and each client device independently selects and requests a most suitable stream for playing according to its own network bandwidth and hardware decoding capability, so as to ensure the smoothness of playing.

[0003] However, in actual application, the prior art has the following problems. Since each client device makes an independent decision, the stream selected by the client device has a huge difference in bit rate and picture quality, which causes the overall visual perception of the spliced screen to be seriously fragmented, and the visual experience is poor. More importantly, the time required for decoding video streams of different bit rates is different, and the difference in network conditions also causes the time of data packets to be inconsistent, which makes it difficult to maintain frame-level precision synchronous playing between devices, and problems such as audio and video out of synchronization and screen tearing often occur. Therefore, the prior art cannot simultaneously guarantee the synchronization and picture consistency of multi-screen display in a heterogeneous network and hardware environment, and thus needs to be improved. SUMMARY

[0004] In order to make up for the above shortcomings, the present application provides a high-definition video multi-screen synchronous transmission system and method, which aims to improve the problem that it is difficult to simultaneously guarantee the synchronization and picture consistency of multi-screen display in a heterogeneous network and hardware environment.

[0005] In a first aspect, the present application provides the following technical solution. A high-definition video multi-screen synchronous transmission system comprises the following modules: A state collection and decision module is configured to collect real-time network states and hardware performance parameters of multiple client devices, determine a system synchronization baseline based on a device with the maximum expected decoding delay, and generate a unified display timestamp sequence; A video preprocessing module is configured to process a video source based on the synchronization baseline, generate a reference synchronous stream for decoding by all client devices, and generate corresponding picture enhancement data for client devices whose network states and hardware performance parameters are better than those of the device corresponding to the system synchronization baseline; A data distribution module is configured to distribute the reference synchronous stream, picture enhancement data and unified display timestamp sequence to each client device; a client-side synchronous rendering module configured to decode the reference synchronous stream according to the received data, wherein the client devices with performance up to the standard incorporate the quality enhancement data and all the client devices are synchronously rendered and displayed according to the uniform display timestamp sequence.

[0006] a dynamic adjustment module configured to periodically or based on an event trigger, re-collect the state parameters of the client devices, and when the total delay of any client device changes by more than a preset threshold, dynamically update the system synchronization baseline and the uniform display timestamp sequence.

[0007] Preferably, the determination process of the system synchronization baseline comprises: estimating the decoding delay of each client device based on the collected hardware performance parameters of the client device; estimating the network transmission delay of each client device based on the collected real-time network status of the client device; adding the decoding delay and the network transmission delay of each client device to obtain the total delay of the client device; comparing the total delays of all the client devices, and determining the total delay with the largest value as the system synchronization baseline.

[0008] Preferably, the generation process of the uniform display timestamp sequence comprises: obtaining a reference time provided by a network time protocol; calculating a theoretical display time point of each video frame relative to the reference time according to the video frame rate; adding the theoretical display time point of each frame and the system synchronization baseline to generate the final display timestamp sequence.

[0009] Preferably, the generation process of the reference synchronous stream comprises: determining the video encoding parameters that can be smoothly decoded by the client device with the lowest performance according to the system synchronization baseline; based on the determined video encoding parameters, transcoding the original video source to output a low-bitrate video stream; using the low-bitrate video stream as the reference synchronous stream for forced decoding by all the client devices.

[0010] Preferably, the generation process of the quality enhancement data comprises: encoding the original video source to generate a high-bitrate high-definition video stream; calculating the pixel data difference between the corresponding frames of the high-definition video stream and the reference synchronous stream; compressing the difference data to generate quality enhancement data.

[0011] Preferably, the distribution process of the reference synchronization stream, the quality enhancement data and the uniform display timestamp sequence comprises: distributing the reference synchronization stream to all client devices through an independent streaming channel; distributing the corresponding quality enhancement data to the client devices meeting the performance requirement through an independent data channel or multiplexing the streaming channel; distributing the uniform display timestamp sequence to all client devices through a signaling channel or encapsulating in the media stream header.

[0012] Preferably, the decoding process of the reference synchronization stream comprises: all client devices receive and parse the transmission data packet of the reference synchronization stream; sending the parsed video encoding data into a local hardware decoder; the hardware decoder completes the decoding operation of the video frame and outputs the raw pixel data.

[0013] Preferably, the fusion process of the quality enhancement data comprises: the client devices meeting the performance requirement receive and decode the quality enhancement data; superimposing the decoded quality enhancement data and the raw pixel data of the corresponding frame obtained by decoding the reference synchronization stream; outputting the superimposed pixel data as the final picture to be rendered.

[0014] Preferably, the synchronization rendering and display process comprises: sending the picture frame to be rendered obtained by decoding and fusion into a rendering queue; comparing the uniform display timestamp sequence with the local system clock of the device; when the local system clock reaches the display timestamp corresponding to a frame, taking out the frame from the rendering queue and submitting to the display buffer for display.

[0015] In a second aspect, the application provides the following technical solution, a high-definition video multi-screen synchronization transmission method, the method comprising the following steps: collecting real-time network status and hardware performance parameters of a plurality of client devices, taking the device with the maximum expected decoding delay as a reference, determining a system synchronization baseline, and generating a uniform display timestamp sequence; based on the synchronization baseline, processing a video source to generate a reference synchronization stream for decoding by all client devices, and generating corresponding quality enhancement data for the client devices meeting the performance requirement; distributing the reference synchronization stream, the quality enhancement data and the uniform display timestamp sequence to each client device; According to the received data, the reference synchronization stream is decoded, wherein the client equipment meeting the performance standard fuses the picture quality enhancement data, and all the client equipment is synchronously rendered and displayed according to the uniform display timestamp sequence. Periodically or based on event triggering, the state parameters of the client equipment are re-collected, and when the total delay of any client equipment changes by more than a preset threshold, the system synchronization baseline and the uniform display timestamp sequence are dynamically updated.

[0016] The present application has the following beneficial effects: 1、In the present application, all equipment decodes the uniform reference stream to ensure synchronization, and the equipment meeting the performance standard further improves the picture quality through enhancement data, and through the strategy of uniform basic picture style and local intelligent enhancement, the visual style of all users is coordinated and consistent under a heterogeneous network, and the overall visual effect of multi-screen splicing display is improved.

[0017] 2、In the present application, the server takes the performance of the equipment with the maximum expected decoding delay as a reference to uniformly plan the video processing and distribution strategy, so as to ensure that the decoding and rendering time axes of all client equipment are strictly consistent, and realize frame-level precision synchronous playback under a heterogeneous hardware environment, and completely eliminate the picture tearing and accumulated delay problems caused by the performance difference of equipment.

[0018] 3、In the present application, the state of the equipment is periodically monitored, and when the network or performance fluctuation exceeds a threshold, the synchronization baseline and timestamp sequence can be dynamically updated, so that the system can not only cope with the difference in the initial state, but also continuously maintain synchronization and optimal picture quality in a long-time running or complex network environment, greatly improving the excellence, stability and user experience of the system. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 The system architecture diagram provided by the present application is shown in the figure. Figure 2 The method flowchart provided by the present application is shown in the figure. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0021] Embodiment one: In the first embodiment of the present application, the present application provides a high-definition video multi-screen synchronous transmission system, as shown in the figure, comprising the following modules: Figure 1 ​A state collection and decision module is configured to collect real-time network states and hardware performance parameters of a plurality of client devices, determine a system synchronization baseline based on a client device with the maximum decoding delay, and generate a unified display timestamp sequence.

[0022] Further, the determination process of the system synchronization baseline comprises: estimating the decoding delay of each client device based on the collected hardware performance parameters; estimating the network transmission delay of each client device based on the collected real-time network states; adding the decoding delay and the network transmission delay of each client device to obtain a total delay of each client device; comparing the total delays of all client devices, and determining the maximum total delay as the system synchronization baseline.

[0023] Further, the generation process of the unified display timestamp sequence comprises: obtaining a reference time provided by a network time protocol; calculating a theoretical display time point of each video frame relative to the reference time according to a video frame rate; adding the theoretical display time point of each frame and the system synchronization baseline to generate a final display timestamp sequence.

[0024] Specifically, the SDK built in each client device periodically sends a heartbeat packet or a special network probe packet to the coordination server. The server evaluates the real-time network quality of the client by calculating the round-trip delay of the data packet and combining the packet loss rate. Finally, the server calculates and maintains an estimated network transmission delay value for each client. When a client first accesses the system, its SDK calls the system API to obtain the hardware information of the device and reports it. These information at least includes: decoder type, such as whether it supports H.265 / HEVC hardware decoding, the maximum resolution and frame rate it supports, and CPU model and core number. The server maintains a device performance database to map the typical single-frame video decoding delay baseline value of the device based on the reported device model; For each client device i, its total delay TotalDelay-i is calculated by the following formula: TotalDelay-i = decode delay-i + network transmission delay-i, wherein decode delay-i is queried from the device performance database, and network transmission delay-i is obtained according to real-time network detection; the server compares the total delay TotalDelay-i of all online client devices, and then selects the total delay with the largest value as a system synchronization baseline of the entire system, which means that the playing rhythm of the system is planned according to the device with the slowest response and the worst network condition in expectation, thereby fundamentally ensuring that all devices can follow the playing rhythm; The coordination server synchronizes with a time server through a network time protocol (NTP) or a more accurate precise time protocol (PTP) to obtain a high-precision and unified reference time, and calculates a theoretical playing time of each frame relative to the reference time according to a frame rate of the video source, for example, the theoretical playing time of the 0th frame is T0, the theoretical playing time of the 1st frame is T0+33.3 ms, the theoretical playing time of the 2nd frame is T0+66.6 ms, and so on. To ensure that the slowest device also has sufficient time to receive and decode data, the above system synchronization baseline is added as a fixed offset to the theoretical display time point of each frame, that is, a final display timestamp = theoretical display time point + system synchronization baseline. A series of absolute time points generated in this way constitute a unified display timestamp sequence, which will be distributed to all clients as a rendering schedule that they must strictly follow. Through the above steps, the processing capacity of the device with the worst expected performance is taken as a global reference, and the rendering time axis of all devices is uniformly planned, thereby ensuring synchronized playing at a frame level and effectively eliminating the problems of screen tearing and audio-visual asynchronization caused by device performance differences.

[0025] The video preprocessing module is configured to process the video source based on the synchronization baseline to generate a reference synchronous stream for decoding by all client devices, and generate corresponding quality enhancement data for client devices whose network status and hardware performance parameters are better than those corresponding to the system synchronization baseline.

[0026] Further, the generation process of the reference synchronous stream includes: According to the system synchronization baseline, determine the video encoding parameters that can be smoothly decoded by the client device with the lowest performance; Based on the determined video encoding parameters, transcode the original video source to output a low-code-rate video stream; The low-code-rate video stream is used as a reference synchronous stream for forced decoding by all client devices.

[0027] Further, the generation process of the quality enhancement data includes: Encode the original video source to generate a high-code-rate high-definition video stream; Calculate the pixel data difference value of the corresponding frame of the high-definition video stream and the reference synchronization stream; Compress the difference value data to generate quality enhancement data.

[0028] Specifically, the video preprocessing module determines the video encoding parameters of the reference stream according to the ability of the client device with the lowest performance corresponding to the system synchronization baseline. Specifically, the video preprocessing module queries the video encoding format supported by the device, the maximum supported resolution, and the highest code rate that can be decoded smoothly. These parameters will be set as the generation target of the reference stream. The video preprocessing module uses an open source library or a hardware encoder to perform real-time transcoding on the original high-definition video source. The transcoding process strictly follows the encoding parameters determined in the previous step. The resolution of the video is scaled to the target size, and the output code rate is controlled within the target code rate range. After transcoding, a low-code rate and low-computational complexity video stream is obtained. This stream is named as the reference synchronization stream and is required to be received and decoded by all client devices in the system; The video preprocessing module uses another encoder to perform high-quality encoding on the same original video source. This process uses a higher target code rate and original resolution to generate a high-code rate high-definition video stream as a quality reference baseline. The video preprocessing module operates at the frame level. For the same video frame at the same time, it obtains two frames of images from the high-definition reference stream and the reference synchronization stream, respectively. Then, through pixel-level operation, the pixel data difference value between the high-definition frame and the reference frame is calculated. This difference image contains the detailed information missing in the reference stream. Since the amount of raw difference image data is still large, the video preprocessing module uses an efficient lossless or visually lossless compression algorithm, such as PNG or lightweight video encoding, to compress the difference data. The compressed data block is the quality enhancement data. These data will strictly correspond to the reference synchronization stream and will only be distributed to client devices that meet the performance requirements; Through the above steps, the reference synchronization stream and the quality enhancement data can be generated, which realizes intelligent hierarchical improvement of quality on the basis of ensuring multi-screen synchronization.

[0029] The data distribution module is used to distribute the reference synchronization stream, the quality enhancement data, and the unified display timestamp sequence to each client device.

[0030] Further, the distribution process of the reference synchronization stream, the quality enhancement data, and the unified display timestamp sequence includes: Distribute the reference synchronization stream to all client devices through an independent streaming media channel; Distribute the corresponding quality enhancement data to client devices that meet the performance requirements through an independent data channel or a multiplexed streaming media channel; The uniform display timestamp sequence is distributed to all client devices through a signaling channel or encapsulated in a media stream header.

[0031] Specifically, the data distribution module distributes the reference synchronization stream to all client devices through an independent streaming media channel. Specifically, the server establishes an independent RTP session or WebRTC media channel for each client, and sends reference synchronization stream data packets through these channels. The use of standard streaming media protocols ensures the real-time and timing of the data. All clients, regardless of performance, receive and decode the same video stream as the synchronization basis through this channel. The data distribution module distributes the quality enhancement data to client devices that meet the performance requirements through an independent data channel. This channel can use WebSocket or a custom application layer protocol based on UDP to establish. Another optional implementation is to multiplex the streaming media channel, such as packaging and sending the quality enhancement data as part of the extension data or payload in the RTP protocol together with the video frames of the reference stream. The server will only send enhancement data packets to IP addresses that meet the conditions according to the maintained device performance list. For the distribution of the uniform display timestamp sequence, a signaling channel or media stream encapsulation can be used. The signaling channel specifically refers to distributing the uniform display timestamp sequence to all client devices through an independent and reliable signaling channel, such as a WebSocket connection based on TCP. The server can send the timestamp list of the entire segment at once or push the timestamp information of the next few seconds in real time during playback. Media stream encapsulation specifically refers to directly encapsulating the timestamp sequence into the media stream. Specifically, the display timestamp corresponding to each frame is inserted as metadata into the RTP packet header extension item of the corresponding video frame or placed in the specified field of the MPEG-TS stream. The client can directly extract the rendering time instruction of the frame while demultiplexing and analyzing the video stream. Through the above steps, efficient and reliable transmission of synchronization data and enhancement data is achieved. The independent streaming media channel ensures stable and low-latency distribution of the reference synchronization stream to all clients, ensuring synchronization. The on-demand data channel transmits quality enhancement data, optimizing network resource utilization. The signaling or encapsulation method issues uniform timestamps, accurately directing the rendering timing of all devices, thereby ensuring the reliability of the final synchronization effect and quality improvement in complex network environments.

[0032] The client synchronization rendering module is configured to decode the reference synchronization stream according to the received data. The client devices that meet the performance requirements fuse the quality enhancement data, and all client devices perform synchronized rendering and display according to the uniform display timestamp sequence.

[0033] Further, the decoding process of the reference synchronization stream includes: All client devices receive and parse the transmission data packets of the reference synchronization stream; The parsed video encoding data is sent to the local hardware decoder; The decoding operation of the video frame is completed by the hardware decoder, and the raw pixel data is output.

[0034] Further, the fusion process of the quality enhancement data includes: The performance-meeting client device receives and decodes the quality enhancement data; The decoded quality enhancement data is superimposed with the raw pixel data of the corresponding frame obtained by decoding the reference synchronization stream; The superimposed pixel data is output as the final picture to be rendered.

[0035] Further, the process of synchronous rendering and display includes: The decoded and fused picture frame to be rendered is sent to the rendering queue; The uniform display timestamp sequence is compared with the local system clock of the device; When the local system clock reaches the display timestamp corresponding to a frame, the frame is taken out from the rendering queue and submitted to the display buffer for display.

[0036] Specifically, the client continuously receives the reference synchronization stream data packets from the server through the network interface. After receiving, the client first performs unpacking and demultiplexing operations to parse the standard video encoding data. The parsed video encoding data is sent to the hardware decoder of the device, such as the MediaCodec of Android, to perform decoding using special hardware to ensure stable and controllable decoding delay. After the hardware decoder completes the decoding, YUV or RGB format raw pixel data is output. At this time, the client obtains the basic picture of the current frame; After completing the decoding of the reference stream, the performance-meeting client device performs the following processes in parallel to improve the picture quality. The client device simultaneously receives and parses the quality enhancement data through the corresponding decoder. The data is compressed and the decoded difference image pixel data is obtained. The client performs addition superposition operation on the difference pixel obtained by decoding the enhancement data and the corresponding pixel of the basic picture obtained by decoding the reference synchronization stream. This operation can restore the missing details and textures in the reference picture. After superposition, a final picture with higher definition and details is generated and output as the picture to be rendered. All client devices, whether they display the base picture or the enhanced picture, must follow the following synchronization process: the decoded and fused picture frame to be rendered is placed in a rendering queue according to its frame number, and the client maintains a local clock synchronized with the server (calibrated by NTP), which continuously compares the local clock with the uniform display timestamp sequence issued from the server. When the local system clock reaches a certain frame, such as the Nth frame corresponding to the accurate display timestamp, the display system takes the Nth frame from the rendering queue and immediately submits it to the display buffer, and the next refresh of the screen will display this frame. Through the above steps, the forced decoding and intelligent fusion mechanism can be implemented on the client, and combined with high-precision synchronization rendering control, the ultimate synchronization and picture quality optimization between heterogeneous devices are ensured.

[0037] The dynamic adjustment module is used to periodically or event-triggeredly re-collect the state parameters of the client devices, and when the total delay of any client device changes by more than a preset threshold, the system synchronization baseline and the uniform display timestamp sequence are dynamically updated.

[0038] Specifically, the dynamic adjustment module is built-in with a timer, such as automatically performing a comprehensive state re-evaluation process every 5 seconds. This periodic check can capture slow and continuous changes in network bandwidth or device load. In addition to periodic checks, the dynamic adjustment module also listens to specific events. When the network transmission delay or decoder load reported by any client device suddenly changes (for example, changes by more than a preset threshold, such as 100 milliseconds, within 1 second), a re-evaluation is triggered immediately. In this way, sudden network jitter or instantaneous fluctuations in device performance can be quickly responded to. The dynamic adjustment module instructs the state collection and decision module to re-collect the latest network status and hardware performance parameters of all online clients, and according to the above process, the total delay of each device is recalculated. The newly calculated total delay set is compared with the current system synchronization baseline. The decision logic is as follows: if the new total delay of all devices has not changed significantly, the existing baseline and timestamp sequence are maintained and do not need to be adjusted. If the total delay of any device changes by more than a preset threshold, or the newly calculated maximum total delay is different from the current system synchronization baseline, it is determined that the baseline needs to be updated. When the decision result is that the update is needed, the dynamic adjustment module sets the new maximum total delay as a new system synchronization baseline, and then it instructs the state collection and decision module to generate a new set of uniform display timestamp sequences based on the new baseline and the uniform reference time, and the dynamic adjustment module broadcasts a notification of a synchronization strategy update to all client devices through a signaling channel, the notification containing brief information of the new reference or the starting point of the new timestamp sequence, and after receiving the notification, the client will smoothly switch to the new timestamp sequence after playing the current frame group being rendered, thereby completing the dynamic adjustment and avoiding dramatic jumps or stalls in the playing process. Through the above steps, the dynamic adjustment module updates the synchronization baseline and the timestamp sequence in real time when the state of any client changes beyond the threshold through the periodic monitoring and the event triggering mechanism, thereby ensuring that the system always maintains frame-level synchronization among multiple screens in a complex environment, and significantly improving the robustness and long-term stability of the system.

[0039] Embodiment Two Since each client device makes an independent decision, the selected stream has a huge difference in code rate and image quality, which causes the overall visual perception of the spliced screen to be seriously fragmented, and the visual experience is poor. More importantly, the time required to decode video streams of different code rates is different, and the difference in network conditions also causes the arrival time of data packets to be inconsistent, which makes it difficult to maintain frame-level synchronization between devices. Therefore, the prior art cannot simultaneously guarantee the synchronization and quality consistency of multi-screen display in a heterogeneous network and hardware environment, and thus needs to be improved. To solve the above problems, the present application provides a high-definition video multi-screen synchronization transmission method, as shown in the structure of Figure 2 The specific implementation process of the method is as follows: First, collect the real-time states of all client devices, determine the system synchronization baseline based on the device with the maximum expected decoding delay, and generate a uniform display timestamp sequence. Then, based on the baseline, process the video source to generate a reference synchronization stream that all devices must decode, and generate quality enhancement data for high-performance devices in parallel. Next, distribute the reference stream, enhancement data, and uniform timestamp to each device through a multi-channel coordination mechanism. All clients then decode the reference stream, with high-performance devices fusing enhancement data to improve image quality and strictly following the uniform timestamp for synchronized rendering and display. Finally, the system continuously monitors the device state through a dynamic adjustment module, and updates the synchronization baseline and timestamp in real time when the delay changes beyond the threshold, ensuring continuous synchronization and stability of the system in a complex environment.

[0040] Finally, it should be noted that the above only describes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art will appreciate that the technical solutions described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalent features, and any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A high-definition video multi-screen synchronous transmission system, characterized in that, Includes the following modules: The status collection and decision-making module is used to collect the real-time network status and hardware performance parameters of multiple client devices, determine the system synchronization baseline based on the device with the highest expected decoding latency, and generate a unified display timestamp sequence. The video preprocessing module is used to process the video source based on the synchronization baseline, generate a reference synchronization stream for decoding by all client devices, and generate corresponding image quality enhancement data for client devices whose network status and hardware performance parameters are better than those of the system synchronization baseline. The data distribution module is used to distribute the baseline synchronization stream, image quality enhancement data, and the unified display timestamp sequence to each client device; The client-side synchronous rendering module is used to decode the baseline synchronous stream based on the received data. Among them, client devices that meet the performance standards integrate the image quality enhancement data, and all client devices perform synchronous rendering and display according to the unified display timestamp sequence. The dynamic adjustment module is used to periodically or based on event triggers to re-collect the status parameters of client devices. When the total latency change of any client device exceeds a preset threshold, the system synchronization baseline and the unified display timestamp sequence are dynamically updated.

2. The high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The process for determining the system synchronization baseline includes: Based on the collected hardware performance parameters of each client device, the decoding latency of a single frame of video is estimated. Based on the real-time network status of each client device collected, its network transmission latency is estimated; The total delay is obtained by adding the decoding delay of each client device to the network transmission delay. The total latency of all client devices is compared, and the total latency with the largest value is determined as the system synchronization baseline.

3. The high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The process for generating the unified display timestamp sequence includes: Obtain the reference time provided by the network clock protocol; Based on the video frame rate, calculate the theoretical display time point of each video frame relative to the reference time; The theoretical display time point of each frame is added to the system synchronization baseline to generate the final display timestamp sequence.

4. The high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The process for generating the reference synchronization stream includes: Based on the system synchronization baseline, determine the video encoding parameters that the client device with the lowest performance can decode smoothly; Based on the determined video encoding parameters, the original video source is transcoded to output a low bitrate video stream. This low-bitrate video stream is used as the baseline synchronization stream for forced decoding by all client devices.

5. A high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The process for generating the image quality enhancement data includes: The original video source is encoded to generate a high-bitrate, high-definition video stream; Calculate the pixel data difference between the corresponding frames of the high-definition video stream and the reference synchronization stream; The difference data is compressed to generate image quality enhancement data.

6. A high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The distribution process for the baseline synchronization stream, image enhancement data, and the unified display timestamp sequence includes: The baseline synchronization stream is distributed to all client devices via a separate streaming media channel; The corresponding image enhancement data is distributed to qualified client devices via an independent data channel or by reusing the streaming media channel. The unified display timestamp sequence is distributed to all client devices via signaling channels or by encapsulation in the media stream header.

7. A high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The decoding process of the reference synchronization stream includes: All client devices receive and parse the transmission data packets of the reference synchronization stream; The parsed video encoded data is sent to the local hardware decoder; The hardware decoder performs the video frame decoding operation and outputs the raw pixel data.

8. A high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The fusion process of the image quality enhancement data includes: A qualified client device receives and decodes the image enhancement data; The decoded image enhancement data is superimposed on the original pixel data of the corresponding frame obtained by decoding the reference synchronous stream; The superimposed pixel data is used as the final output image to be rendered.

9. A high-definition video multi-screen synchronous transmission system according to claim 1, characterized in that, The synchronous rendering and display process includes: The decoded and merged image frames to be rendered are sent to the rendering queue. The unified display timestamp sequence is compared with the device's local system clock; When the local system clock reaches the display timestamp corresponding to a certain frame, the frame is retrieved from the rendering queue and submitted to the display buffer for display.

10. A method for simultaneous multi-screen transmission of high-definition video, characterized in that, For a high-definition video multi-screen synchronous transmission system according to any one of claims 1-9, the method includes the following steps: Collect real-time network status and hardware performance parameters of multiple client devices, determine the system synchronization baseline based on the device with the highest expected decoding latency, and generate a unified display timestamp sequence; Based on the synchronization baseline, the video source is processed to generate a reference synchronization stream for decoding by all client devices, and corresponding image quality enhancement data is generated for client devices that meet the performance requirements. The baseline synchronization stream, image enhancement data, and unified display timestamp sequence are distributed to each client device. Based on the received data, the baseline synchronization stream is decoded, wherein client devices that meet the performance standards fuse the image enhancement data, and all client devices perform synchronous rendering and display according to the unified display timestamp sequence; Periodically or based on event triggers, the status parameters of client devices are re-collected. When the total latency change of any client device exceeds a preset threshold, the system synchronization baseline and the unified display timestamp sequence are dynamically updated.

Images