Video live broadcast method, electronic equipment and storage medium

By obtaining and analyzing live interactive images and integrating video streams, the problem of video anchors being unable to interact when they are busy is solved, the freedom of live broadcast and user stickiness are improved, the synchronous playback of multiple interactive images is achieved, and the variety of live content is enhanced. Sex and interactivity.

CN119071523APending Publication Date: 2024-12-03FUZHOU ROCKCHIP SEMICON

Patent Information

Application Number
CN202411105849.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-13
Publication Date
2024-12-03

AI Technical Summary

Technical Problem

The existing live broadcast equipment has a single function, and video anchors are unable to interact with the audience when they are busy, resulting in a decrease in user stickiness and limited live broadcast time.

Method used

By obtaining and analyzing live interactive images, filtering unless live information, fusion processing to obtain the target interactive images and input videos, generating live video streams, realizing the splicing and synchronous playback of multiple interactive images, and assisting the video anchors in interacting.

Benefits of technology

It improves the freedom of live broadcasts for video anchors, alleviates the pressure of live broadcasts, improves user stickiness and user breadth, and enhances the diversity and interactivity of live broadcast content.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119071523A_ABST
    Figure CN119071523A_ABST
Patent Text Reader

Abstract

The invention discloses a video live broadcast method, electronic equipment and a storage medium. The method comprises the following steps: acquiring a preset background video and at least one live broadcast interaction picture; analyzing and processing the at least one live broadcast interaction picture according to the acquisition source of the at least one live broadcast interaction picture to obtain a target interaction picture corresponding to the live broadcast interaction picture; performing fusion processing on the target interaction picture and the background video to obtain a live video stream; and the live broadcast server sends the live broadcast video stream to a live broadcast audience terminal for live broadcast. The auxiliary live broadcast content can be generated when the video anchor is busy, so that the basic live broadcast interaction requirement is met, and the live broadcast interaction pressure of the video anchor is relieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of video processing technology, and in particular to a video live broadcast method, an electronic device and a storage medium. Background Art

[0002] The commercial ecology and business model of online live broadcast and live shopping are becoming increasingly mature, and video live broadcast has become the mainstream channel for consumers to entertain and shop. The live broadcast industry has had a significant impact on the supply and consumption sides of the business. During live broadcast, the video anchor needs to configure a complete set of live broadcast equipment, then connect to the commercial live broadcast platform, and then provide live broadcast services to consumers. However, the current live broadcast equipment is highly homogenized and has a single function. During the live broadcast process, the video anchor must be next to the live broadcast equipment and can only interact with the audience through sound or picture.

[0003] The current live interactive interface requires the video host to interact with the audience. When the video host is busy, he cannot interact with the audience, and user stickiness decreases. In order to ensure user stickiness, the video host needs to interact with the audience in real time, which limits the live broadcast time of the video host and puts a lot of pressure on live interaction. Summary of the invention

[0004] The present invention provides a video live broadcast method, an electronic device and a storage medium, which can generate auxiliary live broadcast content when a video host is busy to meet basic live broadcast interaction needs, thereby alleviating the live broadcast interaction pressure of the video host.

[0005] In one aspect of the present invention, a method for live video broadcasting is provided. The method comprises: obtaining an input video and at least one live interactive screen; parsing the at least one live interactive screen to obtain a target interactive screen corresponding to the live interactive screen; fusing the target interactive screen with the input video to obtain a live video stream; and sending the live video stream to a live server.

[0006] In another aspect of the present invention, an electronic device is provided, which includes: a memory configured to store an executable program; and one or more processors configured to execute the program so that the electronic device implements the above-mentioned video live broadcast method.

[0007] In another aspect of the present invention, a computer-readable medium is provided, wherein a computer program is stored on the medium, and the computer program is executed by a processor to implement the above-mentioned video live broadcast method.

[0008] According to the technical solution of the present invention, the video anchor can obtain at least one live interactive screen and an input video, and parse at least one live interactive screen to obtain a target interactive screen, thereby filtering out non-live information in the live interactive screen, and ensuring that the core content of the live interactive screen can be displayed. After obtaining the target interactive screen, all target interactive screens and input videos are fused to obtain a live video stream, thereby realizing the splicing and fusion of multiple target interactive screens. Based on this, the video anchor can customize the selection of multiple different types of interactive screens for fusion and simultaneous live display, and assist the video anchor in live interaction when the anchor is busy, meet basic live interaction needs, thereby improving the live broadcast freedom of the video anchor, effectively alleviating the live broadcast pressure of the video anchor, and the synchronous playback of multiple interactive screens can effectively improve user breadth and user stickiness. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A flowchart of a video live broadcast method according to an embodiment of the present invention;

[0010] Figure 2 A flowchart of another video live broadcast method according to an embodiment of the present invention;

[0011] Figure 3 A flowchart of another video live broadcast method according to an embodiment of the present invention;

[0012] Figure 4 A schematic diagram of a fused video of another video live broadcast method according to an embodiment of the present invention;

[0013] Figure 5 The present invention is a schematic structural diagram of a video live broadcast device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0014] In order to explain the technical content, achieved objectives and effects of the present invention in detail, the following is an explanation in combination with the implementation modes and the accompanying drawings.

[0015] In the existing technology, during live broadcast, the video anchor needs to configure a complete set of live broadcast equipment, then connect to the commercial live broadcast platform, and then provide live broadcast services to consumers. However, the current live broadcast equipment is highly homogenized and has single functions. During the live broadcast, the video anchor must be next to the live broadcast equipment and can interact with the audience through sound or picture. The current live broadcast interactive interface requires the video anchor to interact with the audience. When the video anchor is busy, he cannot interact with the audience, and user stickiness decreases. In order to ensure user stickiness, the video anchor needs to interact with the audience in real time, which limits the live broadcast time of the video anchor and puts a lot of pressure on live broadcast interaction.

[0016] In order to solve at least the above technical problems, the present disclosure provides a method for live video broadcasting. According to the present disclosure, at least one live interactive screen and an input video can be obtained, and at the same time, at least one live interactive screen is parsed and processed according to the acquisition source to obtain a target interactive screen, thereby filtering out non-live information in the live interactive screen, and ensuring that the core content of the live interactive screen can be displayed. After obtaining the target interactive screen, all target interactive screens and the input video are fused to obtain a live video stream, thereby realizing the splicing and fusion of multiple target interactive screens. In this way, according to the embodiment of the present disclosure, it is possible to customize the selection of multiple different types of interactive screens for fusion and simultaneous live display, and assist the video anchor to perform live interaction when the anchor is busy, meet the basic live interaction needs, thereby improving the live broadcast freedom of the video anchor, effectively alleviating the live broadcast pressure of the video anchor, and the synchronous playback of multiple interactive screens can effectively improve user breadth and user stickiness.

[0017] Hereinafter, the technical solution according to the present disclosure will be described with reference to specific embodiments and in conjunction with the accompanying drawings.

[0018] Figure 1 1 is a flow chart showing a video live broadcast method 100 according to an embodiment of the present disclosure. Figure 1 The method 100 includes the following steps 102 to 108.

[0019] In step 102, an input video and at least one live interactive screen are obtained.

[0020] In step 104, the at least one live interactive screen is parsed to obtain a target interactive screen corresponding to the live interactive screen.

[0021] In some embodiments, parsing the at least one live interactive screen to obtain a target interactive screen corresponding to the live interactive screen includes: parsing the first live interactive screen including the captured anchor to obtain a first target interactive screen corresponding to the anchor's character information. In some embodiments, additionally or alternatively, parsing the at least one live interactive screen to obtain a target interactive screen corresponding to the live interactive screen includes: parsing the second live interactive screen including the chat interactive window to obtain a second target interactive screen including the automatic reply information in the chat interactive window. In this way, one or more different types of interactive screens are customized and selected for parsing respectively to determine the key live content in the interactive screen and filter out non-live screens.

[0022] In step 106, the target interactive screen is fused with the input video to obtain a live video stream.

[0023] In some embodiments, fusing the target interactive screen with the input video to obtain a live video stream includes: fusing the target interactive screen with the input video through a preset scene fusion and splicing manager to obtain a fused video; encoding the fused video into a compressed video with a predetermined format as the live video stream. In this way, video encoding with a predetermined format has significant advantages such as compression efficiency and high-quality images, and can transmit higher-quality video streams while ensuring transmission efficiency.

[0024] In some embodiments, fusing the target interactive screen with the input video to obtain a live video stream includes: fusing the first target interactive screen corresponding to the anchor character information and / or the second target interactive screen including the automatic reply information in the chat interaction window, and the input background video to obtain the live video stream. In this way, multiple different types of interactive screens are customized and selected for fusion and then displayed live at the same time, meeting the live interactive needs of different users and alleviating the live broadcast pressure of the video anchor.

[0025] In step 108, the live video stream is sent to a live server.

[0026] In some embodiments, sending the live video stream to the live server includes: transmitting the live video stream to the live server via the UVC protocol through the anchor client, so that the live server broadcasts live to the live audience based on the live video stream. In this way, the UVC protocol implements the "plug and play" function of the video device. The user only needs to connect the video device that generates the live video stream to the USB interface of the device running the anchor client, and the anchor client can automatically identify and enable the video device without complicated settings or installation processes, which greatly improves the convenience of using the video device. In addition, the UVC protocol supports the video device to transmit data and power at the same time through the USB cable, without the need for an additional power cable, making the plug-in and unplugging and carrying of the video device more convenient, effectively improving the convenience of the anchor's live broadcast.

[0027] According to an embodiment of the present invention, at least one live interactive screen may include a first live interactive screen and / or a second live interactive screen, as described above.

[0028] In some embodiments, obtaining at least one live interactive screen may include: obtaining a first live interactive screen associated with the anchor from a camera device. At this time, parsing and processing the at least one live interactive screen to obtain a target interactive screen corresponding to the live interactive screen includes: parsing the first live interactive screen based on a preset character recognition algorithm to obtain the anchor character information; and adaptively cropping the first live interactive screen according to the anchor character information to obtain the first target interactive screen. In this way, cropping the anchor screen captured by the camera device can retain the portrait screen of the video anchor to the greatest extent, filter out the non-live screen in the video, so that the video anchor can perform live broadcast tasks in any environment without restricting the live broadcast environment, effectively improving the live broadcast freedom of the video anchor. At the same time, by cropping non-essential live screens so that they can be merged with the input video later, when the video anchor is broadcasting live, the background video can be used to assist the anchor in broadcasting other content.

[0029] In some embodiments, obtaining at least one live interactive screen may include: obtaining a second live interactive screen associated with the chat interactive window from the local device. At this time, parsing the at least one live interactive screen to obtain a target interactive screen corresponding to the live interactive screen includes: parsing the chat interactive window activated in the second live interactive screen; and generating a second target interactive screen based on the chat interactive window. In this way, when the video anchor is unable to broadcast the screen live, it is possible to interact with the live audience end with interactive needs based on the chat interactive window, and only retain the interactive screen, thereby forming a virtual anchor, allowing the video anchor to take an offline rest and relieve the live broadcast pressure of the video anchor. At the same time, when the video anchor is busy and unable to interact with the live audience end, the chat interactive window can also assist the video anchor to interact with the live audience end to improve user stickiness.

[0030] In some embodiments, generating the second target interactive screen based on the chat interaction window includes: performing intelligent interaction based on the text content of the chat interaction window to generate the second target interactive screen including the automatic reply information in the chat interaction window. In this way, the chat interaction window can be automatically replied without manual reply, avoiding the situation where the video anchor is incompatible with text interaction and character interaction, while improving the interaction rate with the live audience during the live broadcast, indirectly increasing the stickiness of the live audience.

[0031] In some embodiments, intelligent interaction based on the text content of the chat interaction window includes: identifying the audience terminal identifier in the chat interaction window, and determining the question information of the audience terminal identifier in the chat interaction window; obtaining the standard reply information corresponding to the question information through the preset chat program interface; querying the interaction record corresponding to the audience terminal identifier; calculating the interaction weight value of the audience terminal identifier according to the interaction record; if the interaction weight value exceeds the preset interaction authority value, reconstructing the standard reply information according to the text style of the target anchor and the interaction record to obtain the anchor reply information; and sending the anchor reply information to the live audience terminal corresponding to the audience terminal identifier to reply to the question information. In this way, it is possible to respond to the audience with a large number of interactions during the live broadcast in a targeted manner. For the audience terminal with more frequent interaction records, the interaction weight value of the audience is higher, that is, the audience stickiness is higher, so the response to the audience adopts a language that is more in line with the anchor style, while improving the interactivity of the live broadcast, ensuring the stickiness of the live audience.

[0032] In some embodiments, the method may further include: if the interaction weight value does not exceed the interaction authority value, sending the standard reply information to the live viewer terminal corresponding to the viewer terminal identifier to reply to the question information. In this way, for a viewer terminal with fewer interaction records, the interaction weight value of the viewer is lower, so the viewer's reply is directly automatically replied with the standard reply information to ensure basic live interaction needs.

[0033] Figure 2 2 is a flow chart showing a video live broadcast method 200 according to an embodiment of the present invention. Figure 2 The method 200 involves obtaining a first live interactive screen associated with the anchor from a camera device, which includes the following steps 202 to 210.

[0034] In step 202, the first live interactive screen of the anchor captured by the camera device is obtained, and the input video is obtained as the preset background screen of the current live screen.

[0035] In some embodiments, the corresponding HDMI-IN picture is obtained from the HDMI interface device as the input video, HDMI is a fully digital video and sound transmission interface; or the corresponding projection picture is obtained from the projection device as the input video.

[0036] In step 204, the region of the video anchor in the current first live interactive screen is determined by a neural network recognition algorithm, that is, the portrait region in the current live interactive screen is identified to obtain the anchor screen.

[0037] In step 206, the anchor screen is adaptively cropped from the first live interactive screen to obtain a target interactive screen.

[0038] In some embodiments, adaptive cropping is performed through an intelligent cropping algorithm to automatically adjust the cropping area according to the host's position and movements to ensure that the picture is always focused on the host.

[0039] In step 208, the anchor screen and the HDMI-IN screen or the projection screen are fused through the scene fusion and stitching manager to obtain a fused video, and the fused video is encoded into a compressed video in H264 format or H265 format as a live video stream.

[0040] In some embodiments, the scene fusion splicing manager needs to support the following performance to achieve fusion processing: 1) Support the simultaneous access of multiple video devices, such as DVD (Digital Video Disc), camera, satellite receiver, set-top box, standard computer, etc. 2) Support the access and processing of multiple signal sources, such as network signal, RGB signal, video signal, etc. 3) Support dynamic switching of signals, complete the allocation and dynamic switching of signals through built-in modules or external modules, and be able to flexibly select and switch different signal sources. 4) Support multi-screen splicing, be able to reorganize the image of the signal input source, reconstruct the image according to the splicing mode selected by the user and output it to the logical display screen or physical display screen. 5) Support edge fusion or edge superposition animation effects to achieve seamless splicing effects. Typical splicing modes applicable to anchor screen fusion include: background video template (exclusive / horizontal two-split screen / vertical two-split screen / horizontal three-split screen / vertical three-split screen / four-split screen / N split screen / asymmetric N split screen) + anchor screen foreground.

[0041] In step 210, the live video stream is transmitted to the live server via the UVC protocol through the anchor client, so that the live server broadcasts live to the live audience terminal based on the live video stream.

[0042] Figure 3 3 is a flow chart showing a video live broadcast method 300 according to an embodiment of the present invention. Figure 3 The method 300 involves obtaining a second live interactive screen associated with the chat interactive window from a local device, which includes the following steps 302 to 318.

[0043] In step 302, the local device plays a preset input video and simultaneously starts a designated chat interaction window.

[0044] In some embodiments, the chat interaction window includes a comment area and a bullet screen area of ​​a live broadcast application, as well as a chat interface of an instant messaging social software.

[0045] In step 304, the local device obtains and recognizes the text content and the viewer terminal identifier in the chat interaction window, and determines the question information of the viewer terminal identifier in the chat interaction window.

[0046] In some embodiments, the viewer terminal identifier is a unique ID number of the viewer terminal.

[0047] In step 306, standard reply information corresponding to the question information is obtained through the chat program interface.

[0048] In some embodiments, a large language model, such as chatGPT or other models, is connected based on a chat program interface, so as to intelligently reply to the question information from the audience based on the large language model.

[0049] In step 308, the interaction record corresponding to the viewer terminal identifier is queried, and the interaction weight value of the viewer terminal identifier is calculated according to the interaction record.

[0050] In some embodiments, the interaction record includes information such as the consumed content, the number of consumption times, and the historical chat records corresponding to the viewer terminal identifier.

[0051] In some embodiments, different interaction records are weighted or voted to calculate the interaction weight value corresponding to the current viewer terminal identifier.

[0052] In step 310, if the interaction weight value exceeds the preset interaction authority value, the standard reply information is reconstructed according to the target anchor's writing style and interaction record to obtain anchor reply information, and the anchor reply information is sent to the live audience terminal corresponding to the audience terminal identifier to reply to the question information.

[0053] In some embodiments, the host's reply message is spoken content.

[0054] In step 312, if the interaction weight value does not exceed the interaction authority value, the standard reply information is directly sent to the live broadcast viewer terminal corresponding to the viewer terminal identifier to reply to the question information.

[0055] In step 314, the chat interaction window currently activated by the local device is analyzed in real time, the chat interaction window is marked as the target interaction screen, and the input video currently played by the local device is marked as the background video.

[0056] In step 316, the auxiliary live video and the chat interaction window are fused through the scene fusion and stitching manager to obtain a fused video, and the fused video is encoded into a compressed video in H264 format or H265 format as a live video stream.

[0057] In some embodiments, the scene fusion and stitching manager can use the chat interaction window as a video input signal source and perform scene fusion and stitching. The camera device runs the scene fusion and stitching manager, and the camera device captures the chat interaction window of the connected mobile phone or tablet device.

[0058] In step 318, the live video stream is transmitted to the live server via the UVC protocol through the anchor client, so that the live server broadcasts live to the live audience terminal based on the live video stream.

[0059] It should be noted that the photographic equipment in the above method is an electronic device such as an IPC camera, mobile phone, set-top box, computer, etc. that includes a video acquisition function and supports UVC protocol transmission. The local device is an electronic device used to execute the above method. The anchor client is an application running on the local device to realize the live broadcast function. The live broadcast audience terminal is an application running on the electronic device to realize the live broadcast function.

[0060] Figure 4 FIG. 4 is a schematic diagram of a fused video according to a video live broadcast method 400 according to an embodiment of the present invention. Figure 4 The method 400 involves simultaneously acquiring a first live interactive screen and a second live interactive screen, and based on the above-mentioned method 200 and method 300, obtaining a first target interactive screen corresponding to the first live interactive screen and a second target interactive screen corresponding to the second live interactive screen, and fusing the first target interactive screen, the second target interactive screen and the input video to obtain a live video stream. Figure 4 402 is the first target interactive screen, 404 is the second target interactive screen, and 406 is the preset input video.

[0061] According to another aspect of the present invention, Figure 5 is a block diagram showing an electronic device 500 according to an embodiment of the present invention. Figure 5 The electronic device 500 includes a memory 502 and one or more processors 504. The memory 502 is configured to store an executable program. The one or more processors 504 are configured to execute the program so that the electronic device implements the above-mentioned video live broadcast method.

[0062] According to another aspect of the present invention, a computer-readable medium is provided, wherein a computer program is stored on the computer-readable medium, and the computer program is executed by a processor to implement the video live broadcast method described above.

[0063] In summary, according to the video live broadcast method, electronic device and storage medium provided by the present invention, the anchor screen captured by the camera device and the display interface collected by the live broadcast device can be simultaneously obtained as the live broadcast interactive screen, and the live broadcast interactive screen is filtered out of the non-live broadcast screen, and then merged with the preset input video to obtain a live video stream, so that multiple contents can be played synchronously under one live broadcast screen, which improves the freedom of live broadcast content and no longer relies on a single live broadcast content, effectively improving user stickiness and user breadth. At the same time, when the live broadcast interactive screen is a chat interactive window, it can automatically reply to the audience's question information based on the interactive record between the video anchor and the audience, and can effectively assist the video anchor in live broadcast interaction when the anchor is busy, reducing the interactive pressure of the video anchor.

[0064] The above descriptions are merely embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent transformations made using the contents of the present invention's specification and drawings, or directly or indirectly applied in related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A video live broadcast method, characterized in that: include: Obtain input video and at least one live interactive screen; Parsing the at least one live interactive screen to obtain a target interactive screen corresponding to the live interactive screen; The target interactive screen is merged with the input video to obtain a live video stream; as well as The live video stream is sent to a live server.

2. The video live broadcast method according to claim 1, characterized in that: Parsing the at least one live interactive screen to obtain a target interactive screen corresponding to the live interactive screen includes: Analyze and process the first live interactive screen including the captured anchor to obtain a first target interactive screen corresponding to the anchor character information; and / or The second live interactive screen including the chat interactive window is parsed and processed to obtain a second target interactive screen including the automatic reply information in the chat interactive window.

3. The video live broadcast method according to claim 2, characterized in that: The target interactive screen is merged with the input video to obtain a live video stream, including: The first target interactive screen corresponding to the anchor character information and / or the second target interactive screen including the automatic reply information in the chat interactive window, and the input video are merged to obtain the live video stream.

4. The video live broadcast method according to claim 1, characterized in that: Acquiring at least one live interactive screen includes: acquiring a first live interactive screen associated with the anchor from a camera device, The at least one live interactive screen is parsed to obtain a target interactive screen corresponding to the live interactive screen, including: Analyzing the first live interactive screen based on a preset character recognition algorithm to obtain the anchor character information; and The first live interactive screen is adaptively cropped according to the anchor character information to obtain a first target interactive screen.

5. The video live broadcast method according to claim 1, characterized in that: Acquiring at least one live interactive screen includes: acquiring a second live interactive screen associated with the chat interactive window from the local device, The at least one live interactive screen is parsed to obtain a target interactive screen corresponding to the live interactive screen, including: parsing the chat interaction window activated in the second live interactive screen; and A second target interaction screen is generated based on the chat interaction window.

6. The video live broadcast method according to claim 5, characterized in that: Generating a second target interactive screen based on the chat interactive window includes: Intelligent interaction is performed based on the text content of the chat interaction window to generate the second target interaction screen including the automatic reply information in the chat interaction window.

7. The video live broadcast method according to claim 6, characterized in that: Intelligent interaction based on the text content of the chat interaction window includes: Identify the audience terminal identifier in the chat interaction window, and determine the question information of the audience terminal identifier in the chat interaction window; Obtaining standard reply information corresponding to the question information through a preset chat program interface; Querying the interaction record corresponding to the viewer terminal identifier; Calculate the interaction weight value of the audience terminal identifier according to the interaction record; If the interaction weight value exceeds the preset interaction authority value, the standard reply information is reconstructed according to the target anchor's writing style and the interaction record to obtain anchor reply information; and The anchor reply information is sent to the live broadcast audience terminal corresponding to the audience terminal identifier to reply to the question information.

8. The video live broadcast method according to claim 7, characterized in that: Also includes: If the interaction weight value does not exceed the interaction authority value, the standard reply information is directly sent to the live broadcast viewer terminal corresponding to the viewer terminal identifier to reply to the question information.

9. The video live broadcast method according to claim 1, characterized in that: The target interactive screen is merged with the input video to obtain a live video stream, including: fusing the target interactive screen with the input video through a preset scene fusion and splicing manager to obtain a fused video; and The fused video is encoded into a compressed video with a predetermined format as the live video stream.

10. The video live broadcast method according to claim 1, characterized in that: Sending the live video stream to a live streaming server includes: The live video stream is transmitted to the live server via the UVC protocol through the anchor client, so that the live server broadcasts live to the live audience terminal based on the live video stream.

11. An electronic device, characterized in that: include: a memory configured to store an executable program; as well as One or more processors are configured to execute the program so that the electronic device implements the video live broadcast method according to any one of claims 1 to 10.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: The computer program is executed to implement the video live broadcast method according to any one of claims 1 to 10.

Citation Information

Patent Citations

  • Bullet screen reply method, bullet screen reply device and live broadcast equipment

    CN110035325A

  • Live video generation method and device, computer readable medium and electronic equipment

    CN111432235A

  • Live broadcast method and device based on a virtual anchor, equipment and storage medium

    CN113766253A

  • Method for superimposing live image of person onto real scene, and electronic device

    WO2021218547A1

Cited By

  • Live video architecture for commentator

    TWI929919B