Low-bandwidth high-definition video conferencing system and video transmission method based on the financial industry

By using video conferencing terminals and video servers in the financial industry video conferencing system, only the head posture and facial expression information of video participants are transmitted, and image reconstruction is used using AI technology, the problem of poor video quality under low bandwidth is solved. Real-time high-definition video conferencing is realized, reducing bandwidth requirements, and is a low-cost solution suitable for the financial industry.

CN118509616BActive Publication Date: 2024-11-05EASTERN COMM
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410954079.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-05
Estimated Expiration
2044-07-16

AI Technical Summary

Technical Problem

The remote video banking in the financial industry has poor video quality under low bandwidth conditions, resulting in poor user experience. The existing solutions can achieve video transmission by sacrificing image quality, which cannot meet the needs of application scenarios.

Method used

Using a combination of a video conferencing terminal and a video server, it uses the head posture information and facial expression information of the video participants to compress and encode, only feature data is transmitted, and image reconstruction is carried out on the video receiving end to realize high-definition video conferencing. Reduce data transmission volume and adapt to low bandwidth environments.

Benefits of technology

Without renovating the bank network infrastructure, real-time high-definition video conferencing is achieved, reducing bandwidth requirements by 50%, providing low-cost solutions and improving user experience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118509616B_ABST
    Figure CN118509616B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of Internet video conferencing, and discloses a low-bandwidth high-definition video conferencing system and a video transmission method based on the financial industry. The video sending end of the system obtains the initial image of the video participant after entering the video conference and the head posture and facial expression information of each frame, compresses and uniquely ID codes and sends them to the video server; the video server caches the information sent by the video sending end and the corresponding ID code; the initial image and its ID code are transmitted to the video receiving end, and the ID code information corresponding to the head posture and facial expression information is compared with the cached ID code. When there is inconsistency, the head posture and facial expression information and the corresponding ID code are sent to the video receiving end; the video receiving end performs AI video reconstruction on the decompressed initial image, head posture and facial expression information. The present invention realizes a real-time high-definition video conferencing system without the need to transform the original network infrastructure of the bank through minimized data transmission.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of Internet video conferencing, and in particular to a low-bandwidth high-definition video conferencing system and a video transmission method based on the financial industry. Background Art

[0002] In recent years, with the rapid development of the Internet, Internet-based video conferencing systems have been qualitatively improved. However, due to various limitations such as the infrastructure of the internal network, the cost of hardware upgrades, and low bandwidth, remote video banking has not been fully promoted in the financial industry's internal network. The transmission of high-definition video under low bandwidth has always been problematic. The current popular solution is the bit rate adaptive mode, which sacrifices image quality to achieve video transmission, reducing user experience. This method cannot meet existing application scenarios. Summary of the invention

[0003] In view of this, the present invention provides a low-bandwidth high-definition video conferencing system and a video transmission method based on the financial industry to solve the problem of poor video quality in remote video banking under low bandwidth conditions, resulting in poor user experience.

[0004] In a first aspect, the present invention provides a low-bandwidth high-definition video conferencing system based on the financial industry, comprising: a video conferencing terminal and a video server, wherein the video conferencing terminal comprises a user terminal and at least one bank agent terminal for conducting a video conference with the user terminal, wherein both the user terminal and the bank agent terminal can serve as a video sending terminal or a video receiving terminal, wherein:

[0005] The video sending end is used to obtain the initial image of the video participant after entering the video conference and the head posture information and facial expression information of each frame of the video participant in the video conference, and compress the initial image, head posture information and facial expression information and send them to the video server after unique ID encoding;

[0006] The video server is used to cache the initial image and its ID code sent by the video sending end and the head posture information and facial expression information of each frame of the video participant within a preset time period and the corresponding ID code; and transmit the initial image and its ID code to the video receiving end, and traverse and compare the ID code information corresponding to the head posture information and facial expression information sent by the video sending end in real time with the cached ID code, and when they are inconsistent, send the head posture information and facial expression information and the corresponding ID code to the video receiving end;

[0007] The video receiving end is used to receive and decompress the information sent by the video server, and to perform AI video reconstruction based on the decompressed initial image, head posture information and facial expression information.

[0008] The low-bandwidth high-definition video conferencing system based on the financial industry provided in this embodiment sends the original image of the video participant only once, only collects the head posture information and facial expression information of each frame, and encodes it. When transmitting the collected feature data, first compare whether the data has been transmitted within the preset time period. If it is inconsistent, it is transmitted. The data that has been transmitted is directly extracted from the local cache. If it is not obtained, the latest feature data is pulled again, thereby minimizing data transmission, and using AI technology at the video receiving end to drive the initial appearance feature value information image to synthesize the facial replay result through head posture and facial expression. The embodiment of the present invention realizes a real-time high-definition video conferencing system without the need to transform the original network infrastructure of the bank, balancing the benefits and cost maximization, and providing a feasible solution for the subsequent development of video banks.

[0009] In a second aspect, the present invention provides a video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry, which is applied to a video conferencing terminal, comprising:

[0010] Obtaining the initial image of the video participant after entering the video conference and the head posture information and facial expression information of each frame of the video participant in the video conference, and compressing the initial image, head posture information and facial expression information and sending them to the video server after corresponding unique ID encoding;

[0011] Receive and decompress the information sent by the video server, and perform AI video reconstruction based on the decompressed initial image, head posture information, and facial expression information.

[0012] The embodiments of the present invention reduce the amount of data transmission by collecting the head posture information and facial expression information of the initial image and each frame of the image, compressing and transmitting them, and performing video reconstruction after decompression. High-definition video conferencing is reconstructed with AI to achieve high-definition and low-code, which can reduce the bandwidth by an average of 50%, providing the financial industry with a low-cost solution to reduce costs and increase efficiency.

[0013] In an optional implementation, the method further includes: when it is detected that there is an obstruction on the face of a video participant or a new object is added to the video background, the initial image is updated, compressed, and uniquely ID-encoded before being sent to the video server.

[0014] The embodiment of the present invention monitors the changes in the background of the characters, timely reacquires a new image to update the initial image when entering the video conference, and then reconstructs the video based on the new source image.

[0015] In an optional implementation, the AI ​​video reconstruction based on the decompressed initial image, head posture information, and facial expression information includes:

[0016] Input the decompressed facial expression information and head posture information into a preset three-dimensional face reconstruction model to obtain a corresponding feature point image;

[0017] Rendering synthesizes facial reenactment video based on initial image and landmark image.

[0018] The embodiment of the present invention is different from the face replacement method or the facial expression capture and animation method. The appearance of the character in the reconstructed video comes from the source image, and its action comes from the character action in the driving video, which can reduce the bandwidth by 50% on average, thereby achieving low-bandwidth, high-quality video reconstruction.

[0019] In a third aspect, a video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry is applied to a video server, comprising:

[0020] Cache the initial image and its ID code sent by the video transmitter and the head posture information and facial expression information of each frame of the video participant within a preset time period and the corresponding ID code;

[0021] The initial image and its ID code are transmitted to the video receiving end, and the ID code information corresponding to the head posture information and facial expression information sent by the video sending end in real time is traversed and compared with the cached ID code. When there is inconsistency, the head posture information and facial expression information ID code and the corresponding one are sent to the video receiving end.

[0022] The video server in the embodiment of the present invention implements a real-time high-definition video conferencing system by minimizing data forwarding without modifying the original network infrastructure of the bank, thereby providing a feasible solution for the subsequent development of video banks.

[0023] In a fourth aspect, the present invention provides a video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry, which is applied to a video conferencing terminal and includes:

[0024] The video sending module is used to obtain the initial image of the video participant after entering the video conference and the head posture information and facial expression information of each frame of the video participant in the video conference, and compress the initial image, head posture information and facial expression information and send them to the video server after corresponding unique ID encoding;

[0025] The video receiving module is used to receive and decompress the information sent by the video server, and perform AI video reconstruction based on the decoded information.

[0026] In a fifth aspect, the present invention provides a video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry, which is applied to a video server and includes:

[0027] A local cache module is used to cache the initial image and its ID code sent by the video transmitter and the head posture information and facial expression information of each frame of the video participant within a preset time period and the corresponding ID code;

[0028] The information sending module is used to transmit the initial image and its ID code to the video receiving end, and to traverse and compare the ID code information corresponding to the head posture information and facial expression information sent by the real-time video sending end with the cached ID code. When there is inconsistency, the head posture information and facial expression information ID code and the corresponding one are sent to the video receiving end.

[0029] In a sixth aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to execute the video transmission method based on the low-bandwidth high-definition video conferencing system for the financial industry according to the second aspect, the third aspect or any corresponding embodiment thereof.

[0030] In the seventh aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute the video transmission method based on a low-bandwidth high-definition video conferencing system for the financial industry according to the second aspect, the third aspect or any corresponding embodiment thereof.

[0031] In an eighth aspect, the present invention provides a computer program product, comprising computer instructions, wherein the computer instructions are used to enable a computer to execute the video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry according to the second aspect, the third aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0033] Figure 1 is a schematic diagram of the structure of a low-bandwidth high-definition video conferencing system based on the financial industry according to an embodiment of the present invention;

[0034] Figure 2 is an interactive schematic diagram of a low-bandwidth high-definition video conferencing system for the financial industry according to an embodiment of the present invention;

[0035] Figure 3It is a flow chart of a video transmission method applied to a video conferencing terminal of a low-bandwidth high-definition video conferencing system in the financial industry according to an embodiment of the present invention;

[0036] Figure 4 It is a flow chart of a video transmission method applied to a video server of a low-bandwidth high-definition video conferencing system in the financial industry according to an embodiment of the present invention;

[0037] Figure 5 It is a structural block diagram of a video transmission device applied to a video conferencing terminal of a low-bandwidth high-definition video conferencing system in the financial industry according to an embodiment of the present invention;

[0038] Figure 6 is a structural block diagram of a video transmission device applied to a video server of a low-bandwidth high-definition video conferencing system in the financial industry according to an embodiment of the present invention;

[0039] Figure 7 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution 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, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0041] According to an embodiment of the present invention, there is provided an embodiment of a video transmission method based on a low-bandwidth high-definition video conferencing system for the financial industry. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer device such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0042] The present invention is aimed at the particularity of remote video conferencing in the financial and banking industries. Under normal circumstances, there is only one user in each signal channel, and the background is also static. Each frame only needs to transmit head posture information and facial expression information, and the amount of data transmitted is small. The video is further compressed, and the AI ​​video reconstruction method based on deep learning solves the problem of insufficient bandwidth of traditional video conferencing technology. The video transmission method based on the low-bandwidth high-definition video conferencing system in the financial industry provided by the embodiment of the present invention reduces the high bandwidth requirements under high-definition video due to the small amount of data transmitted over the network, and at the same time truly restores the user's micro-expressions and movements, meeting the needs of realizing high-definition video conferencing under low-bandwidth conditions in the office environment of the financial industry.

[0043] In this embodiment, a low-bandwidth high-definition video conferencing system based on the financial industry is provided. Figure 1 As shown, it includes: a video conferencing terminal and a video server. The video conferencing terminal includes a user terminal and at least one bank agent terminal for video conferencing with the user terminal. The user terminal and the bank agent terminal can both be used as a video sending terminal or a video receiving terminal. The specific process of the interaction between the three is as follows Figure 2 As shown:

[0044] The video sending end is used to obtain the initial image of the video participant after entering the video conference and the head posture information and facial expression information of the video participant in each frame of the video conference, and compress the initial image, head posture information and facial expression information and send them to the video server after unique ID encoding.

[0045] The video server is used to cache the initial image and its ID code sent by the video sending end and the head posture information and facial expression information and the corresponding ID code of each frame of the video participant within a preset time period; and transmit the initial image and its ID code to the video receiving end, and traverse and compare the ID code information corresponding to the head posture information and facial expression information sent by the video sending end in real time with the cached ID code, and when there is inconsistency, send the head posture information and facial expression information and the corresponding ID code to the video receiving end;

[0046] The video receiving end is used to receive and decompress the information sent by the video server, and perform AI video reconstruction based on the decompressed initial image, head posture information and facial expression information.

[0047] In the embodiment of the present invention, the video participants are users and bank agents, and the user end and the bank agent end having a video conference with the user end are one-to-one or one-to-many video conference scenarios (one user having a video conference with one or more bank agents). The user end can use a mobile phone equipped with a camera, a CRM deposit and withdrawal machine, a smart tablet, etc. to shoot the user in real time, and the bank agent end can obtain the information of the bank agent through the camera device in the bank agent system, and obtain the initial image of the user or bank agent in the video from the real-time communication. The video server records the video sender, video receiver, and dynamic data, and then forwards it to one or more video receivers.

[0048] The data encoding format in the embodiment of the present invention adopts entropy coding. Each frame only needs to transmit compressed head posture information and facial expression information. The data to be transmitted is relatively small. Overall, compared with the H264 format, it can save ten times the network bandwidth.

[0049] In the embodiment of the present invention, the initial image is sent only once, and the video sender stores the feature data of each frame sent within half an hour in a local loop on the video server. If the data to be sent is consistent with the local data and is within half an hour, only the unique ID code is sent to the video server to inform the server that the feature data already exists on the server. The video server extracts data from the stored feature value record library and forwards it to the video receiving end.

[0050] At the video receiving end, AI technology is used to drive the initial user appearance feature value information image synthesis facial replay results through the head posture and facial expression data transmitted over the network, truly restoring the user's micro-expressions and movements, meeting the needs of the financial industry to achieve high-definition video conferencing under low-bandwidth office conditions.

[0051] The low-bandwidth high-definition video conferencing system based on the financial industry provided by the embodiment of the present invention only sends the original image of the video participant once, only collects the head posture information and facial expression information of each frame, and encodes it. When transmitting the collected feature data, first compare whether the data has been transmitted within the preset time period. If it is inconsistent, it is transmitted. The data that has been transmitted is directly taken from the local cache. If it is not taken, the latest feature data is pulled again, thereby minimizing data transmission, and using AI technology at the video receiving end to drive the initial user appearance feature value information image synthesis face replay result through the two data of head posture and facial expression. The embodiment of the present invention realizes a real-time high-definition video conferencing system without the need to transform the original network infrastructure of the bank, balances the benefits and maximizes the cost, and provides a feasible solution for the subsequent development of video banks.

[0052] In this embodiment, a video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry is provided, which is applied to a video conferencing terminal. Figure 3 is a flow chart of a video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry according to an embodiment of the present invention. Figure 3 As shown, the process includes the following steps:

[0053] Step S301, obtaining the initial image of the video participant after entering the video conference and the head posture information and facial expression information of each frame of the video participant in the video conference, and compressing the initial image, head posture information and facial expression information and sending them to the video server after corresponding unique ID encoding.

[0054] Specifically, the video participants in the embodiment of the present invention are users and bank agents. Their initial images are collected only once after entering the video conference for unique ID coding and identification. The face boundary frame is captured from each frame of the video conference to obtain the face frame facial contour and key feature points such as the eyes, nose, and mouth. The face calibration method is a relatively mature face recognition system, such as the real-time face recognition and monitoring system (IntraFace). Using cascaded face key point positioning, 49 key points can be accurately predicted to identify facial expressions. A deep learning-based method (such as using deep learning models such as convolutional neural networks (CNN) or recurrent neural networks (RNN)) directly identifies head posture information from face images. Entropy coding is used to compress the head posture information and facial expression information, and then uniquely coded and identified and sent to the video server.

[0055] Specifically, the embodiment of the present invention monitors the changes in the background of the characters in real time at the video conference terminal. When it is detected that there are obstructions on the faces of the video participants or new objects are added to the video background, the initial image is updated and compressed and sent to the video server after unique ID encoding. For example, when a new object enters the background of the video, or when the video participants wear glasses, masks, hats and other obstructions, a new image is re-acquired to update the initial image when entering the video conference, and then the video is reconstructed based on this new source image, and this new source image is also uploaded in a compressed manner.

[0056] Step S302, receiving and decompressing the information sent by the video server, and decompressing the information based on the decoded information received by the video receiving end, and performing AI video reconstruction based on the decoded initial image, head posture information and facial expression information.

[0057] The embodiments of the present invention reduce the amount of data transmission by collecting the head posture information and facial expression information of the initial image and each frame of the image, compressing and transmitting them, and performing video reconstruction after decompression. High-definition video conferencing is reconstructed with AI to achieve high-definition and low-code, which can reduce the bandwidth by an average of 50%, providing the financial industry with a low-cost solution to reduce costs and increase efficiency.

[0058] In one embodiment, step S302 specifically includes the following steps:

[0059] Step S3021: inputting the decompressed facial expression information and head posture information into a preset three-dimensional face reconstruction model to obtain a corresponding feature point image;

[0060] Step S3022: Rendering a synthetic facial reenactment video based on the initial image and the feature point image.

[0061] Specifically, the 3DMM algorithm is used to fit the shape of the face in the image (obtained based on the feature points of the initial image), expression and head posture and calculate the corresponding feature point image, and then the fitted feature point image and the face image are input into the rendering network to synthesize the facial reenactment video, which is only an example and not limited to this.

[0062] The embodiment of the present invention is different from the face replacement method or the facial expression capture and animation method. The appearance of the character in the reconstructed video comes from the source image, and its action comes from the character action in the driving video, which can reduce the bandwidth by 50% on average, thereby achieving low-bandwidth, high-quality video reconstruction.

[0063] In this embodiment, a video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry is provided, which is applied to a video server. Figure 4 is a flow chart of a video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry according to an embodiment of the present invention. Figure 4 As shown, the process includes the following steps:

[0064] Step S301, caching the initial image and its ID code sent by the video transmitter and the head posture information and facial expression information of the participants in each frame of video within a preset time period and the corresponding ID code.

[0065] Step S302, transmit the initial image and its ID code to the video receiving end, and compare the ID code information corresponding to the head posture information and facial expression information sent by the real-time video sending end with the cached ID code. When there is inconsistency, the head posture information and facial expression information ID code and the corresponding one are sent to the video receiving end.

[0066] In one embodiment, the video server stores each frame of feature data sent within half an hour in a local loop. If the data to be sent is consistent with the local data and within half an hour, only the unique ID code is sent. The video server extracts data from the stored feature value record library and forwards it to the recipient, realizing a real-time high-definition video conferencing system without the need to transform the original network infrastructure of the bank, providing a feasible solution for the subsequent development of video banks.

[0067] In this embodiment, a video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0068] This embodiment provides a video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry, such as Figure 5 As shown, it is applied to video conferencing terminals, including:

[0069] The video sending module 501 is used to obtain the initial image of the video participant after entering the video conference and the head posture information and facial expression information of each frame of the video participant in the video conference, and compress the initial image, head posture information and facial expression information and send them to the video server after corresponding unique ID encoding;

[0070] The video receiving module 502 is used to receive and decompress information sent by the video server, and perform AI video reconstruction based on the decoded information.

[0071] In some optional implementations, when the video sending module 501 detects that there is an obstruction on the face of a video participant or a new object is added to the video background, the initial image is updated, compressed, and uniquely ID-encoded before being sent to the video server.

[0072] In some optional implementations, the video receiving module 502 includes:

[0073] A three-dimensional face reconstruction unit, used to input the decompressed facial expression information and head posture information into a preset three-dimensional face reconstruction model to obtain a corresponding feature point image;

[0074] The rendering unit is used for rendering and synthesizing the facial reenactment video based on the initial image and the feature point image.

[0075] The embodiment of the present invention also provides a video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry, which is applied to a video server, such as Figure 6 As shown, including:

[0076] The local cache module 601 is used to cache the initial image and its ID code sent by the video transmitter and the head posture information and facial expression information of each frame of the video participant within a preset time period and the corresponding ID code;

[0077] The information sending module 602 is used to transmit the initial image and its ID code to the video receiving end, and to traverse and compare the ID code information corresponding to the head posture information and facial expression information sent by the real-time video sending end with the cached ID code. When there is inconsistency, the head posture information and facial expression information ID code and the corresponding one are sent to the video receiving end.

[0078] The further functional description of each of the above modules and units is the same as that of the above corresponding embodiments and will not be repeated here.

[0079] The video transmission device based on the low-bandwidth high-definition video conferencing system in the financial industry in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0080] The embodiment of the present invention also provides a video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry, having the above Figure 5 or Figure 6 The video transmission device shown is based on a low-bandwidth high-definition video conferencing system for the financial industry.

[0081] See also Figure 7 , Figure 7 is a schematic diagram of the structure of a computer device provided by an optional embodiment of the present invention, such as Figure 7 As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 7 A processor 10 is taken as an example.

[0082] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0083] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0084] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0085] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0086] The computer device further comprises a communication interface 30 for the computer device to communicate with other devices or a communication network.

[0087] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor, or hardware, the method shown in the above embodiment is implemented.

[0088] A part of the present invention may be applied as a computer program product, such as a computer program instruction, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should understand that the existence of the computer program instruction in a computer-readable medium includes, but is not limited to, a source file, an executable file, an installation package file, etc., and accordingly, the way in which the computer program instruction is executed by the computer includes, but is not limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding installed program. Here, the computer-readable medium may be any available computer-readable storage medium or communication medium accessible to the computer.

[0089] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A low-bandwidth high-definition video conferencing system based on the financial industry, characterized in that: include: A video conference terminal and a video server, wherein the video conference terminal includes a user terminal and at least one bank agent terminal for video conferencing with the user terminal, wherein both the user terminal and the bank agent terminal can be used as a video sending terminal or a video receiving terminal, wherein: The video sending end is used to obtain the initial image of the video participant after entering the video conference and the head posture information and facial expression information of each frame of the video participant in the video conference, and compress the initial image, head posture information and facial expression information and send them to the video server after unique ID encoding; The video server is used to cache the initial image and its ID code sent by the video sending end and the head posture information and facial expression information and the corresponding ID code of each frame of video participants within a preset time period; and transmit the initial image and its ID code to the video receiving end, and traverse and compare the ID code information corresponding to the head posture information and facial expression information sent by the video sending end in real time with the ID code of the cache, and when they are inconsistent, send the head posture information and facial expression information and the corresponding ID code to the video receiving end, and when they are consistent, extract data from a stored feature value record library based on the head posture information and facial expression information and the corresponding ID code and forward it to the video receiving end; The video receiving end is used to receive and decompress the information sent by the video server, and to perform AI video reconstruction based on the decompressed initial image, head posture information and facial expression information.

2. A video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry, applied to a video conferencing terminal, characterized in that: include: Obtaining an initial image after a video participant enters a video conference and head posture information and facial expression information of each frame of video participants in the video conference, and compressing the initial image, head posture information and facial expression information and sending them to a video server after corresponding unique ID codes, wherein the video server is used to cache the initial image and its ID code sent by the video sending end and the head posture information and facial expression information of each frame of video participants within a preset time period and the corresponding ID code; The initial image and its ID code are transmitted to the video receiving end, and the ID code information corresponding to the head posture information and facial expression information sent by the video sending end in real time is compared with the cached ID code. When they are inconsistent, the head posture information, facial expression information and the corresponding ID code are sent to the video receiving end. When they are consistent, data is extracted from the stored feature value record library based on the head posture information, facial expression information and the corresponding ID code and forwarded to the video receiving end. Receive and decompress the information sent by the video server, and perform AI video reconstruction based on the decompressed initial image, head posture information, and facial expression information.

3. The method according to claim 2, characterized in that Also includes: When it is detected that there is an obstruction on the face of the video participant or a new object is added to the video background, the initial image is updated and compressed and uniquely ID-encoded before being sent to the video server.

4. The method according to claim 2 or 3, characterized in that: The AI ​​video reconstruction based on the decompressed initial image, head posture information and facial expression information includes: Input the decompressed facial expression information and head posture information into a preset three-dimensional face reconstruction model to obtain a corresponding feature point image; Rendering synthesizes facial reenactment video based on initial image and landmark image.

5. A video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry, applied to a video server, characterized in that: include: Cache the initial image and its ID code sent by the video transmitter and the head posture information and facial expression information of each frame of the video participant within a preset time period and the corresponding ID code; The initial image and its ID code are transmitted to the video receiving end, and the ID code information corresponding to the head posture information and facial expression information sent by the video sending end in real time is traversed and compared with the cached ID code. When they are inconsistent, the head posture information, facial expression information and the corresponding ID code are sent to the video receiving end. When they are consistent, data is extracted from the stored feature value record library based on the head posture information, facial expression information and the corresponding ID code and forwarded to the video receiving end. The video receiving end performs AI video reconstruction on the initial image, head posture information and facial expression information sent by the video server.

6. A video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry, applied to a video conferencing terminal, characterized in that: include: The video sending module is used to obtain the initial image of the video participant after entering the video conference and the head posture information and facial expression information of each frame of the video participant in the video conference, and compress the initial image, head posture information and facial expression information and send them to the video server after corresponding unique ID coding, and the video server is used to cache the initial image and its ID coding sent by the video sending end and the head posture information and facial expression information of each frame of the video participant in a preset time period and the corresponding ID coding; The initial image and its ID code are transmitted to the video receiving end, and the ID code information corresponding to the head posture information and facial expression information sent by the video sending end in real time is compared with the cached ID code. When they are inconsistent, the head posture information, facial expression information and the corresponding ID code are sent to the video receiving end. When they are consistent, data is extracted from the stored feature value record library based on the head posture information, facial expression information and the corresponding ID code and forwarded to the video receiving end. The video receiving module is used to receive and decompress the information sent by the video server, and perform AI video reconstruction based on the decoded information.

7. A video transmission device based on a low-bandwidth high-definition video conferencing system in the financial industry, characterized in that: Applied to a video server, the feature is that it includes: A local cache module is used to cache the initial image and its ID code sent by the video transmitter and the head posture information and facial expression information of each frame of the video participant within a preset time period and the corresponding ID code; The information sending module is used to transmit the initial image and its ID code to the video receiving end, and to traverse and compare the ID code information corresponding to the head posture information and facial expression information sent by the real-time video sending end with the cached ID code. When they are inconsistent, the head posture information, facial expression information and the corresponding ID code are sent to the video receiving end. When they are consistent, data is extracted from the stored feature value record library based on the head posture information, facial expression information and the corresponding ID code and forwarded to the video receiving end. The video receiving end performs AI video reconstruction on the initial image, head posture information and facial expression information sent by the video server.

8. A computer device, characterized in that: include: A memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the video transmission method based on a low-bandwidth high-definition video conferencing system for the financial industry as described in any one of claims 2 to 4 or 5 by executing the computer instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the video transmission method based on a low-bandwidth high-definition video conferencing system for the financial industry as described in any one of claims 2 to 4 or 5.

10. A computer program product, characterized in that It includes computer instructions, which are used to enable a computer to execute the video transmission method based on a low-bandwidth high-definition video conferencing system in the financial industry as described in any one of claims 2 to 4 or 5.