TW202304212A - Live broadcast method, system, computer equipment and computer readable storage medium - Google Patents

Abstract

The present disclosure provides a live broadcast method, system, computer equipment and computer-readable storage medium, wherein the method includes: determining a virtual live broadcast mode; the virtual live broadcast mode is used to indicate at least one target capture area for limb capture of a real anchor; collecting the video image of the real anchor during the live broadcast; based on the virtual live broadcast mode, performing motion recognition on the target capture area in the video image to obtain a target recognition result; based on the target recognition result, determining the video stream data corresponding to the virtual anchor model driven by the real anchor; the video stream data is used to present a process in which the virtual anchor model performs the action indicated by the identification result.

Description

Live broadcast method, system, computer equipment and computer-readable storage medium

The present invention relates to the technical field of computers, and specifically relates to a live broadcast method, system, computer equipment and computer-readable storage media.

In the virtual live broadcast process in the related art, the anchor needs to wear a motion capture device, and then the motion data of the anchor terminal is captured by the motion capture device. Due to the limitation of the motion capture device, the motion capture device in the related art can only capture part of the body movements of the anchor, so that the virtual live broadcast software in the related art cannot meet the motion capture requirements of the anchor. At the same time, the motion capture device limits the live broadcast scene of the anchor, and limits the actions that the live broadcast can make, and increases the complexity of the virtual live broadcast.

Embodiments of the present invention at least provide a live broadcast method, system, computer equipment, and computer-readable storage medium.

In the first aspect, the embodiment of the present invention provides a live broadcast method, which is applied to the anchor terminal device, including: determining a virtual live broadcast mode; the virtual live broadcast mode is used to indicate at least one target capture part for body capture of the real anchor; collecting the Describe the video image of the real anchor in the live broadcast process; based on the virtual live broadcast mode, perform action recognition on the target capture part in the video image to obtain the target recognition result; based on the target recognition result, determine the real The video stream data corresponding to the virtual anchor model driven by the anchor; the video stream data is used to present the process of the virtual anchor model executing the action indicated by the target recognition result.

In the embodiment of the present invention, after the virtual live broadcast mode is determined, the virtual live broadcast mode determines the target capture position, without manually setting the target capture position, and can automatically filter out the body parts that do not need motion capture, thereby realizing automatic The body parts to be captured of the real anchor are captured, thereby shortening the time for motion capture and reducing the complexity of the virtual live broadcast. At the same time, the live broadcast method can also be adapted to live broadcast scenes with high immediacy.

In an optional implementation manner, the determining the virtual live broadcast mode includes: determining the operating scenario of the anchor device; based on multiple operating versions pre-set for the anchor end device, determining the The target running version; determine the virtual live broadcast mode based on the determined target running version.

In the embodiment of the present invention, by determining the above operation scenario, the computing capability of the installation device corresponding to the host device can be determined, and then the running version that meets its computing capability can be run on the installation device, thereby ensuring the normal operation of the host device. Running, reducing the problem of abnormal operation of the host device due to the large computing demand and insufficient computing power of the installed device.

In an optional implementation manner, the determining the running scenario of the host device includes: obtaining the device identification information of the host device when the first installation operation for the live broadcast application is detected, and according to The device identification information determines the running scenario; and/or, when an opening operation for the live broadcast application is detected, obtain the remaining device computing resources of the anchor device at the current moment, and use the remaining The computing resource of the device determines the running scenario.

In the embodiment of the present invention, by detecting the first installation operation or opening operation of the anchor device, the computing capability of the anchor device at the current moment can be determined before the anchor device is turned on, so that before the anchor device is turned on, , accurately determine the running scenario of the host device according to the computing capability, and then ensure the normal operation of the host device.

In an optional implementation manner, the determining the virtual live broadcast mode includes: responding to a live broadcast mode selection instruction of the real host, and determining the virtual live broadcast mode in a plurality of preset live broadcast modes.

In the embodiment of the present invention, by setting the way for the user to select the virtual live broadcast mode of the real anchor in the live broadcast mode, more abundant virtual live broadcast scenes can be provided for the user, thereby meeting the needs of the user and improving the user experience .

In an optional implementation manner, the method further includes: before collecting the video image of the real anchor during the live broadcast, collecting a preview image containing the real anchor; determining the Whether the target capture part meets the action recognition conditions; If the target capture part included in the preview image does not satisfy the motion recognition condition, generate target adjustment information until it is determined that the target capture part satisfies the motion recognition condition, wherein the target adjustment information It is used to remind the real anchor to adjust the display state of the target capturing part in the preview image.

In the embodiment of the present invention, when it is determined based on the preview image that the target capture part does not meet the action recognition conditions, target adjustment information is generated, which can realize the display of the camera and/or the target capture part before the live broadcast by the user The state is adjusted to improve the accuracy of action recognition during the live broadcast, thereby improving the live broadcast effect.

In an optional implementation manner, the performing action recognition on the target capture part in the video image based on the virtual live broadcast mode to obtain the target recognition result includes: detecting that the target capture part contains a hand In the case of a part, obtain the mode tag corresponding to the virtual live broadcast mode, wherein the mode tag includes whether to perform hand recognition on the hand part; if the mode tag is determined to be the target mode tag Next, perform motion recognition on the target capturing part in the video image, and perform hand detection on the hand part in the target capturing part, to obtain the target recognition result including the hand detection result.

In an optional implementation manner, after determining the video stream data corresponding to the virtual anchor model driven by the real anchor, the method further includes: detecting the hand gesture of the real anchor according to the hand detection result as The preset gesture is used to obtain the rendering material special effect corresponding to the preset gesture; and render the rendering material special effect in a specified video frame in the video stream data.

In the embodiment of the present invention, through the above-mentioned processing method, it is possible to perform refined processing on the target capture part, so as to meet different live broadcast requirements of users.

In an optional implementation manner, the determining the video stream data corresponding to the virtual anchor model driven by the real anchor based on the target recognition result includes: obtaining a live viewing request sent by at least one audience device; The live viewing request determines the interface background image of the live viewing interface corresponding to each audience device, wherein the interface background image includes: a static background image or a dynamic background image; it is determined based on the target recognition result for representing The virtual anchor model performs a plurality of video images of the actions indicated by the target recognition result; the background image of each video image in the plurality of video images is replaced by the interface background image, and based on The modified video images determine the video stream material.

In an optional implementation manner, the method further includes: after determining the video stream data corresponding to the virtual anchor model driven by the real anchor, pushing an image containing the interface background image to the audience device corresponding to the same interface background image video stream data.

In the embodiment of the present invention, the corresponding video stream data is generated for the viewer-end devices requesting the same interface background image, and the video stream data including the interface background image is pushed to the viewer-end device requesting the same interface background image , which can save the time for determining the video stream data, so as to improve the live broadcast quality of the virtual live broadcast.

In an optional implementation manner, the method further includes: after determining the virtual live broadcast mode, displaying the first indication information and/or the second indication information in the display interface of the anchor device; wherein, the first The indication information is used to indicate the target capture part in the valid capture state, and the second indication information is used to indicate the target capture part in the invalid capture state.

In the embodiment of the present invention, by displaying the first instruction information and/or the second instruction information on the display interface of the anchor device, the real anchor can be instructed to make effective capture actions, thereby improving the accuracy of motion capture. To ensure the stable operation of the host device.

In a second aspect, an embodiment of the present invention provides a live broadcast system, including: an anchor device and an audience device; the anchor device is configured to determine the The video stream data corresponding to the virtual anchor model driven by the real anchor, and push the video stream data to the viewer device; Describe the video stream data.

In an optional implementation manner, the viewer end device includes: a mobile terminal device and a PC device; in the case that the viewer end device is the mobile terminal device, the anchor end device transmits content (Content Delivery Network, CDN for short) distribution network to transmit the video stream data to the viewer-end device; if the viewer-end device is the PC device, the host device transmits the video stream data to the viewer-end device through the CDN distribution network and the forwarding stream server The viewer device transmits the video stream data.

In a third aspect, an embodiment of the present invention provides a live broadcast device, which is set on the host device, including: a first determination unit configured to determine a virtual live broadcast mode; the virtual live broadcast mode is used to indicate at least A target capture part; the collection unit is configured to collect the video image of the real anchor during the live broadcast; the action recognition unit is configured to perform actions on the target capture part in the video image based on the virtual live broadcast mode Recognition to obtain a target recognition result; a second determination unit configured to determine the video stream data corresponding to the virtual anchor model driven by the real anchor based on the target recognition result; the video stream data is used to present the virtual anchor model The process of executing the action indicated by the recognition result.

In the fourth aspect, the embodiment of the present invention also provides an electronic device, including: a processor, a memory, and a bus bar. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the The processor communicates with the memory through a bus, and when the machine-readable instructions are executed by the processor, the above-mentioned first aspect, or the steps in any possible implementation manner of the first aspect are executed.

In the fifth aspect, the embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned first aspect, or any one of the first aspects steps in a possible implementation.

An embodiment of the present invention also provides a computer program, the computer program includes computer-readable codes, and when the computer-readable codes run in an electronic device, the processor of the electronic device executes the program described in any one of the above-mentioned embodiments. The live broadcast method described above.

In order to make the above-mentioned objects, features and advantages of the present invention more comprehensible, preferred embodiments will be described in detail below together with the accompanying drawings.

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only It is a part of embodiments of the present invention, but not all embodiments. The elements of the embodiments of the invention generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts belong to the protection scope of the present invention.

It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures.

The term "and/or" in this article only describes an association relationship, which means that there may be three relationships, for example, A and/or B may mean: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the term "at least one" herein means any one or any combination of at least two of the plurality, for example, including at least one of A, B, and C, may mean including the composition consisting of A, B, and C Any one or more elements selected in the collection.

After research, it is found that in the process of virtual live broadcast in related technologies, the anchor needs to wear a motion capture device, and then capture the motion data of the anchor end through the motion capture device. Due to the limitation of the motion capture device, the motion capture device in the related art can only capture part of the body movements of the anchor, so that the virtual live broadcast software in the related art cannot meet the motion capture requirements of the anchor. At the same time, the motion capture device limits the live broadcast scene of the anchor, and limits the actions that the live broadcast can make, and increases the complexity of the virtual live broadcast.

Based on the above research, an embodiment of the present invention provides a live broadcast method. The live broadcast method provided by the embodiment of the present invention can be applied in a virtual live broadcast scenario. The virtual live broadcast scene can be understood as the use of pre-set virtual anchor models, such as red pandas, little rabbits, cartoon characters, etc. to replace the actual image of the real anchor for live broadcast. At this time, the above-mentioned virtual anchor models are shown in the live video screen . At the same time, the interaction between the real anchor and the audience can also be carried out according to the virtual anchor model.

For example, the camera device of the anchor device can collect video images containing the real anchor, and capture the body of the real anchor contained in the video image, so as to obtain the posture information of the real anchor. After the posture information is determined, a corresponding driving signal can be generated, and the driving signal is used to drive the animation special effect corresponding to the virtual anchor model displayed in the live video screen.

In an optional embodiment, the real anchor may preset a corresponding virtual anchor model, for example, the preset virtual anchor model may be "YYY role model in XXX game". A real anchor can preset one or more virtual anchor models. When starting the virtual live broadcast at the current moment, one can be selected from one or more preset virtual anchor models as the virtual anchor model at the current moment. Wherein, the virtual anchor model may be a two-dimensional (two Dimensional, 2D for short) model, and may also be a three-dimensional (three Dimensional, 3D for short) model.

In another optional implementation, in addition to determining the virtual anchor model for the real anchor in the manner described above, after the video image is acquired, the virtual anchor model can be reshaped for the real anchor in the video image.

For example, the real anchor included in the video image can be identified, so as to reshape the virtual anchor model for the real anchor according to the recognition result. The recognition result may include at least one of the following: the gender of the real anchor, the appearance characteristics of the real anchor, the clothing characteristics of the real anchor, and the like.

At this point, a model that matches the recognition result can be searched from the virtual anchor model database as the virtual anchor model of the real anchor. For example, according to the recognition result, it is determined that the real anchor wears a peaked cap and clothes in hip-hop style during the live broadcast. At this time, a virtual anchor model that matches the "cap" or "hip-hop style" can be searched from the virtual anchor model database as the virtual anchor model of the real anchor.

In addition to searching for a model that matches the recognition result in the virtual anchor model library, it is also possible to construct a module through the model based on the recognition result to instantly build a corresponding virtual anchor model for the real anchor.

Here, when constructing the virtual anchor model in real time, the virtual anchor model used by the virtual live broadcast initiated by the real anchor in the past can also be used as a reference to construct the virtual anchor model driven by the real anchor at the current moment.

Through the method of determining the virtual anchor model described above, it is possible to customize corresponding virtual anchor models for real anchors, thereby making the virtual anchor models rich and varied. At the same time, by customizing the virtual anchor model, it can also leave a deeper impression on the audience.

In order to facilitate the understanding of this embodiment, a live broadcast method disclosed in the embodiment of the present invention is first introduced in detail. The execution subject of the live broadcast method provided by the embodiment of the present invention is generally a computer device with a certain computing capability, and the computer device can be Devices that support the installation of the above-mentioned host devices. In some possible implementation manners, the live broadcast method may be implemented by a processor calling computer-readable instructions stored in a memory.

Referring to FIG. 1A , which is a flowchart of a live broadcast method provided by an embodiment of the present invention, the method includes steps S101 to S107 as follows.

S101: Determine a virtual live broadcast mode; the virtual live broadcast mode is used to indicate a target capture position for body capture of a real anchor.

In the embodiment of the present invention, the virtual live broadcast mode may include but not limited to: head and shoulders live broadcast mode, waist and waist live broadcast mode, and whole body live broadcast mode.

Here, the head-and-shoulders live broadcast mode can be understood as the action recognition of the parts above the shoulders of the real anchor. The waist live broadcast mode is understood as the action recognition of the parts above the waist of the real anchor. The whole body live broadcast mode can be understood as the action recognition of the whole body limbs of the real anchor.

In addition to the virtual live broadcast mode described above, other live broadcast modes may also be included, which will not be described here as examples.

S103: Collect video images of the real anchor during the live broadcast.

Here, the video image may be captured by a camera installed on the host device where the host device is located.

S105: Based on the virtual live broadcast mode, perform action recognition on the target capture part in the video image to obtain a target recognition result.

Here, the target capture part is the target capture part of the real anchor; the target recognition result includes at least one of the following: the position information of the body key points of each target capture part, the size information of the face detection frame, and the position information of the face detection frame.

If the target capture part includes a hand, the target recognition result also includes at least one of the following: size information of the hand detection frame, position information of the hand detection frame, gesture classification information of the player framed by the hand detection frame, and The validity of the gesture indicated by the gesture classification information.

Here, the validity is used to represent whether the gesture indicated by the gesture classification information satisfies the special effect triggering condition. If so, it is a valid gesture; otherwise, it is an invalid gesture.

S107: Based on the target recognition result, determine the video stream material corresponding to the virtual anchor model driven by the real anchor; the video stream material is used to present a process in which the virtual anchor model performs an action indicated by the recognition result.

In the embodiment of the present invention, after the virtual live broadcast mode is determined, the virtual live broadcast mode determines the target capture position, without manually setting the target capture position, and can automatically filter out the body parts that do not need motion capture, thereby realizing automatic The body parts to be captured of the real anchor are captured, thereby shortening the time for motion capture and reducing the complexity of the virtual live broadcast. At the same time, the live broadcast method can also be adapted to live broadcast scenes with high immediacy.

FIG. 1B shows a schematic diagram of a system architecture to which the live broadcasting method according to an embodiment of the present invention can be applied; In order to support an exemplary application, the video image acquisition terminal 201 and the control terminal 203 establish a communication connection through the network 202, and the video image acquisition terminal 201 reports to the control terminal 203 through the network 202 the collected real host during the live broadcast process. In the video image, the control terminal 203, based on the virtual live broadcast mode, performs action recognition on the target capture part in the video image; and based on the target recognition result, determines the video stream data corresponding to the virtual anchor model driven by the real anchor; finally, the control The terminal 203 displays the video stream data in the live video interface, and sends it to the video image acquisition terminal 201 through the network 202 .

As an example, the video image acquisition terminal 201 may include an image acquisition device, and the control terminal 203 may include a visual processing device with visual information processing capability or a remote server. The network 202 can be connected in a wired or wireless manner. Wherein, when the control terminal 203 is a visual processing device, the video image acquisition terminal 201 can communicate with the visual processing device through a wired connection, such as performing data communication through a bus; when the control terminal 203 is a remote server, The video image acquisition terminal 201 can exchange data with a remote server through a wireless network.

Or, in some scenarios, the video image acquisition terminal 201 may be a vision processing device with a video acquisition module, or a host computer with a camera. At this time, the live broadcasting method of the embodiment of the present invention may be executed by the video image acquisition terminal 201 , and the above-mentioned system architecture may not include the network 202 and the control terminal 203 .

In an optional implementation manner, as shown in FIG. 2, for the above step S101, determining the virtual live broadcast mode includes the following process: Step S11, determining the operation scenario of the anchor device; wherein, the operation scenario is used to represent the hardware computing resources of the installation device corresponding to the anchor device; Step S12: Determine a target running version that matches the running scenario based on multiple running versions pre-set for the anchor device; Step S13, determining the virtual live broadcast mode based on the determined target running version.

In the embodiment of the present invention, firstly, the running scenario of the host device can be determined, and the running scenario can be understood as the installation scenario of the host device. For example, the installation scene of the installation device corresponding to the host device (that is, the above-mentioned computer device). After the operating scenario is determined, a virtual live broadcast mode matching the operating scenario can be determined.

In the embodiment of the present invention, multiple running versions are pre-set for the host device, wherein each running version corresponds to a corresponding running tag, and the running tag is used to indicate the running scenario corresponding to the running version. For different operating versions, different virtual live broadcast modes are preset.

Here, the running version may be the running version of the live broadcast software, and different running versions correspond to different virtual live broadcast modes.

In an optional implementation manner, the running version may be understood as a software development kit (Software Development Kit, SDK for short) version of the live broadcast software.

For example, when installing the live broadcast software on the host device for the first time, you can choose to install the software development kit SDK that matches the running scene of the current host device.

In another optional implementation manner, the running version can also be understood as multiple software running modes in the software development kit SDK of the live broadcast software.

For example, when the live broadcast software is installed on the host device for the first time, the software development kit SDK of the live broadcast software can be installed. When running the live broadcast software on the anchor device, the corresponding operation mode in the live software can be automatically run according to the operation scenario (for example, remaining computing resources) of the anchor device.

After the running scenario is determined, the target running version matching the running scenario can be determined based on the version tags corresponding to the multiple running versions preset for the anchor device. Furthermore, the virtual live broadcast mode of the real anchor is determined based on the determined target running version.

In the embodiment of the present invention, it is assumed that the multiple operating versions described above include operating version 1 and operating version 2, wherein operating version 1 can support driving the following parts: head expression driving, torso driving, arm driving, fingers Drive; running version 2 can support drive the following parts: expression drive and trunk drive.

In the embodiment of the present invention, by determining the above operation scenario, the computing capability of the installation device corresponding to the host device can be determined, and then the running version that meets its computing capability can be run on the installation device, so as to ensure the normal operation of the host device , to reduce the problem of abnormal operation of the host device due to the large computing demand and the insufficient computing power of the installed device.

In an optional implementation manner, the above-mentioned step S11, determining the running scenario of the host device, includes the following methods.

method one: When the first installation operation for the live broadcast application is detected, the device identification information of the host device is obtained, and the running scenario is determined according to the device identification information. In the embodiment of the present invention, when it is detected that the user installs the live application program for the first time, the device identification information of the anchor device can be obtained, and then, according to the device identification information of the anchor device, the calculation of the anchor device can be determined. ability. Afterwards, the running scenario can be determined based on the determined computing capability. For example, the running scenario is an anchor device with strong computing capability, or the running scenario is an anchor device with weak computing capability.

Method 2: When an opening operation for the live broadcast application is detected, the remaining device computing resources of the host device at the current moment are acquired, and the running scenario is determined according to the remaining device computing resources. In the embodiment of the present invention, when it is detected that the user opens the live broadcast application, the remaining device computing resources of the host device can be obtained, and then the host device can be determined according to the remaining device computing resources of the host device. computing power. Afterwards, the running scenario can be determined based on the determined computing capability. For example, the running scenario is an anchor device with strong computing capability, or the running scenario is an anchor device with weak computing capability.

Method 3: In the embodiment of the present invention, when it is detected that the user installs the live broadcast application program for the first time and opens the live broadcast application program, the device identification information of the host device and the remaining device computing resources of the host device can be obtained, and then, According to the device identification information of the host device and the remaining computing resources of the device, the computing capability of the host device is determined. Afterwards, the operating scenario of the host device can be determined according to the determined computing capability. For example, the operating scenario of the anchor device is an anchor device with strong computing capability, or the operating scenario of the anchor device is an anchor device with weak computing capability.

In the above embodiments, by detecting the first installation operation or opening operation, it is possible to determine the computing capability of the anchor device at the current moment before starting the live broadcast application, so that before the anchor device is turned on, it can be accurately calculated according to the calculation capability. Ability to determine the operating scenario of the host device, thereby ensuring the normal operation of the host device.

In an optional implementation manner, for the above step S101, determining the virtual live broadcast mode includes the following steps.

Step S21, in response to the selection instruction of the live broadcast mode of the real anchor, determine the virtual live broadcast mode among a plurality of preset live broadcast modes.

In the embodiment of the present invention, a plurality of preset live broadcast modes may be preset, for example, "head shoulder scene" and "waist scene" as shown in FIG. 3 .

After the user turns on the host device, as shown in FIG. 3 , the plurality of preset live broadcast modes can be displayed on the display interface of the host device. The user can select the corresponding live broadcast mode as the virtual live broadcast mode in the display interface shown in Figure 3, for example, the user can choose to enter the "head and shoulders scene", or choose to enter the "waist scene".

When the user chooses to enter the "head and shoulders scene", the information shown in Figure 4 can be displayed on the display interface; when the user chooses to enter the "waist scene", the information shown in Figure 5 can be displayed on the display interface information.

In the above embodiments, by setting the way for the user to select a virtual live broadcast mode from multiple preset live broadcast modes, more abundant virtual live broadcast scenes can be provided for the user, thereby meeting the needs of the user and improving the user experience.

In an embodiment of the present invention, the method further includes the following steps: After determining the virtual live broadcast mode of the real anchor, displaying the first indication information and/or the second indication information in the display interface of the anchor end device; Wherein, the first indication information is used to indicate the target capture part in the effective capture state, and the second indication information is used to indicate the target capture part in the invalid capture state.

In the embodiment of the present invention, after the user chooses to enter the "head and shoulders scene", the information shown in FIG. 4 can be displayed on the display interface. After the user chooses to enter the "waist scene", the information shown in Figure 5 can be displayed on the display interface. At this time, the first indication information and/or the second indication information may be displayed on the display interface.

Here, the first indication information is used to indicate the target capture part in an effective capture state. For example, in the "head and shoulders scene" mode, the head of the real anchor completely appears in the video image, or the shoulders of the real anchor completely appear in the video image. For example, in this "waist scene" mode, the complete upper body parts of the real anchor appear in the video image.

Here, the quantity of the second indication information may be multiple, or one. The different second indication information is used to indicate target capture parts in different invalid capture states. For example, as shown in FIG. 4 , two second indication information are displayed, and the two second indication information respectively indicate the target capture positions of the anchor in different invalid capture states.

In the above embodiments, by displaying the first instruction information and/or the second instruction information on the display interface of the anchor device, the real anchor can be instructed to make effective capture actions, thereby improving the accuracy of motion capture and ensuring The stable operation of the host device.

In the embodiment of the present invention, after determining the virtual live broadcast mode, and before collecting the video images of the real anchor during the live broadcast, the method further includes the following steps: (1) Collect a preview image containing the real host; (2) Determine whether the target capture part included in the preview image satisfies the motion recognition condition; (3) When the target capture part included in the preview image does not satisfy the motion recognition condition, generate target adjustment information until it is determined that the target capture part satisfies the motion recognition condition, wherein the The target adjustment information is used to remind the real anchor to adjust the display state of the target capturing part in the preview image.

In the embodiment of the present invention, after the user selects the virtual live broadcast mode, the user can start to collect preview images containing real anchors. At this time, you can check whether the target capture part at the current moment meets the action recognition condition through the preview image. In some embodiments, the action recognition condition is a preset action of the target capture part. If the action of the target capture part meets the preset action, it means that the target capture part satisfies the action recognition condition; The preset action indicates that the target capture part does not satisfy the action recognition condition.

When it is determined that the condition is satisfied, the user can be prompted to enter the live broadcast mode. At this time, the user can click the "Enter" button in the preview interface to enter the live broadcast interface corresponding to the virtual live broadcast mode.

If it is determined that the condition is not satisfied, it may be determined that the target capture part in the preview image is in an invalid capture state, for example, as indicated by the second indication information in FIG. 4 or FIG. 5 . At this time, the prompt information "the target capture part is invalid" can be displayed in the display interface of the preview image, and at the same time, target adjustment information can also be generated and displayed, which is used to remind the real anchor to adjust the target capture position installation. For example, the target adjustment information may be the following adjustment information: Please move your body up, or move the camera device down to ensure that your upper body is in the preview image; please adjust your sitting posture, or, please adjust the orientation of the camera device, This ensures that your face is within the preview image.

In the above-mentioned embodiment, when it is determined based on the preview image that the target capture part does not meet the motion recognition conditions, the target adjustment information is generated, which can realize the display status of the photography device and/or the target capture part before the live broadcast by the user. Adjust, so as to improve the accuracy of action recognition during the live broadcast, and then improve the live broadcast effect.

In the embodiment of the present invention, for the above step S105, based on the virtual live broadcast mode, the action recognition is performed on the target capture part in the video image to obtain the target recognition result, including the following steps: Step S1051, when it is detected that the target capture part includes a hand part, obtain the mode tag corresponding to the virtual live broadcast mode, wherein the mode tag includes a target mode tag whether to perform hand recognition on the hand part ; Step S1052, when it is determined that the pattern label is the target pattern label, perform motion recognition on the target capture part in the video image, and perform hand detection on the hand part in the target capture part, The target recognition result including the hand detection result is obtained.

In the embodiment of the present invention, assuming that the target capture part indicated by the virtual live broadcast mode includes hand parts, then after the real anchor enters the virtual live broadcast mode, finger actuation information can be displayed on the virtual live broadcast interface of the anchor end, for example, the The finger actuation information may be information as shown in FIG. 6 . As shown in Figure 6, the start buttons of "finger-driven" and "gesture recognition" can be displayed in the virtual live broadcast interface of the anchor. When the real host chooses to enable "finger drive" and/or "gesture recognition", a pattern label for recognizing hand parts can be correspondingly generated.

Therefore, in the embodiment of the present invention, when it is determined according to the virtual live broadcast mode that the target capture part includes hand parts, the mode tag corresponding to the virtual live broadcast mode may also be acquired. According to the mode label, it can be determined whether the real anchor has turned on "finger drive" and "gesture recognition". When it is determined that the mode label is the target mode label, it is determined that at least one of "finger-driven" and "gesture recognition" is turned on. At this time, it can also be determined based on the target mode label that the real anchor has turned on "finger-driven" , or "Gesture Recognition" is turned on.

For example, if the target mode label is "01", it is determined that "Finger Drive" is turned on; if the target mode label is "10", it is determined that "Gesture Recognition" is turned on; if the target mode label is "11", it is determined that it is turned on "finger drive" and "gesture recognition".

In the embodiment of the present invention, when the "finger driving" is turned on, the hand detection result includes position information of each key point of the hand. When “Gesture Recognition” is turned on, the hand detection result includes the gesture recognition result. For example, the gesture recognition result can be that the hand is in an OK posture, the hand is in a heart-to-heart posture, and so on. When "Finger Drive" and "Gesture Recognition" are turned on, the hand detection result includes the gesture recognition result and the position information of each key point of the hand.

As shown in FIG. 7 , gestures that can be recognized by the host device, such as "OK", "666" and "hello", can also be displayed in the presentation interface of the host terminal.

In the above-mentioned embodiment, through the above-mentioned processing method, it is possible to perform refined processing on the target capture part, so as to meet different live broadcast requirements of users.

In the embodiment of the present invention, after determining the video stream data corresponding to the virtual anchor model driven by the real anchor, the method further includes the following steps: (1) According to the hand detection result, it is detected that the hand gesture of the real anchor is a preset gesture, and the rendering material special effect corresponding to the preset gesture is acquired; (2) Render the special effect of the rendering material in a specified video frame in the video stream data.

In the embodiment of the present invention, when the real anchor turns on "gesture recognition", the hand detection result includes the gesture recognition result. At this time, it can be determined according to the gesture recognition result whether the real anchor's hand gesture is Default gestures. If so, determine the rendering material special effect corresponding to the preset gesture, and render the rendering material special effect in the specified video frame in the video stream data.

In an optional implementation manner, before determining that the hand gesture of the real anchor is a preset gesture according to the hand detection result, the following steps may also be performed: Firstly, pose detection is performed on the real anchor in the video image to obtain the pose detection result, and then, according to the body detection result in the pose detection result, it is determined whether the video image meets the gesture recognition condition.

In the embodiment of the present invention, after the limb detection result described above is determined, the relative positional relationship between each target capture part is determined according to the limb detection result; according to the relative positional relationship, it is determined whether the video image satisfies the gesture recognition condition .

It can be understood that the above relative positional relationship includes at least one of the following: a relative distance between each target capturing part, and an angular relationship between associated limb parts in each target capturing part. Wherein, the associated body parts may be understood as adjacent target capture parts, or target capture parts of the same type.

In the embodiment of the present invention, the video image satisfying the gesture recognition condition may be understood as: the body movement of the real host in the video image is a preset body movement.

Secondly, when it is determined that the gesture recognition condition is satisfied, determine whether the hand gesture of the real anchor is a preset gesture according to the hand detection result; and when the preset gesture is detected, in the video Add the rendering material special effect to the specified video frame in the streaming data.

In the embodiment of the present invention, the rendering material special effect corresponding to the video image is determined through the combination of the limb detection result and the hand detection result. In this case, when it is determined that the real anchor meets the gesture recognition conditions according to the gesture detection results, the way of recognizing the gesture of the real anchor can improve the efficiency of gesture comparison and shorten the time of gesture comparison, so that this The technical solution can be applied to live broadcast scenarios with high immediacy requirements.

In the embodiment of the present invention, the above step S107, based on the target recognition result, determines the video stream data corresponding to the virtual anchor model driven by the real anchor, which specifically includes the following steps: Step S1071, obtaining a live viewing request sent by at least one viewer device; Step S1072: Determine the interface background image of the live viewing interface corresponding to each viewer device based on the live viewing request, wherein the interface background image includes: a static background image or a dynamic background image; Step S1073, determining a plurality of video images used to represent that the virtual anchor model performs the action indicated by the target recognition result; Step S1074, replacing the background image of each video image in the plurality of video images with the interface background image, and determining the video stream data based on the modified video images.

In the embodiment of the present invention, before determining the video stream material corresponding to the virtual anchor model driven by the real anchor, a live viewing request sent by at least one viewer terminal device may also be acquired. The live viewing request may carry the interface background image requested by each viewer device. If the viewer device does not request the interface background image, then the interface background image can be set as a transparent background image.

At this point, it is possible to determine a plurality of video images used to characterize the virtual anchor model performing the action indicated by the target recognition result, and replace the background image of each video image in the plurality of video images with the interface background image, and determine the video stream data based on the modified video images.

In the embodiment of the present invention, each video image in a plurality of video images may be segmented into a foreground image and a background image, and then the foreground image and the interface background image obtained by the segmentation are Fusion is performed to obtain multiple modified video images. For each video image, a corresponding time tag (for example, a time stamp) is included. At this time, multiple modified video images may be processed based on the time tag to generate the video data flow.

After the video stream data corresponding to the virtual anchor model to be driven is determined, the video stream data including the interface background image can also be pushed to the audience device corresponding to the same interface background image.

In the above embodiment, the method of generating corresponding video stream data for the viewer-end devices requesting the same interface display background, and pushing the video stream data containing the interface background image to the viewer-end devices requesting the same interface background image can be Save time for determining video stream data to improve live broadcast quality of virtual live broadcast.

In the embodiment of the present invention, the above step: performing action recognition on the target capture part in the video image to obtain the target recognition result, also includes the following process: When it is detected that the target capture part contained in the video image is incomplete, the video image frame can also be expanded to obtain the target image; The detected area; the pose detection is performed on the target image through the pre-trained pose detection model to obtain the pose detection result of the real anchor.

Using the above processing method, it is possible to detect the posture of the real anchor even when the video image frame does not contain complete body parts, so as to ensure the normal and stable operation of the anchor device.

In the embodiment of the present invention, based on the target recognition result, determining the video stream data corresponding to the virtual anchor model driven by the real anchor further includes the following steps: Determine the position information of the target capture part according to the target recognition result, and determine the position information of the key points of the corresponding part according to the driving signal of the corresponding part of the virtual anchor model in the virtual live image through the driving signal. For each video image, a corresponding virtual live image can be correspondingly determined, and then video stream data can be obtained. In this embodiment of the present invention, the virtual anchor model may be a three-dimensional avatar (3D avatar) model.

Referring to FIG. 8 , which is a schematic diagram of a live broadcast system provided by an embodiment of the present invention, the live broadcast system includes: a host device 71 and a viewer device 72 .

The anchor device is configured to determine the video stream data corresponding to the virtual anchor model driven by the real anchor according to the live broadcast method described in any one of the above, and push the video stream data to the viewer end device.

The viewer device is configured to obtain the video stream data, and play the video stream data on a live viewing interface.

Here, the viewer device includes: a mobile terminal device and a PC device.

In the case that the viewer device is the mobile terminal device, the host device transmits the video stream data to the viewer device through a CDN distribution network.

In the case that the viewer-end device is the PC device, the host-end device transmits the video stream data to the viewer-end device through a CDN distribution network and a forwarding stream server.

It can be seen from FIG. 8 that the host device is installed in the host device 71 . First, the virtual live broadcast mode of the real anchor can be determined; after the virtual live broadcast mode is determined, the parts to be driven (or target capture parts) can be determined according to the virtual live broadcast mode, such as expressions, arms and torso waiting to be driven. Afterwards, the video image of the real anchor during the live broadcast can be collected by the camera device 71 of the anchor end device, and based on the virtual live broadcast mode, the action recognition is performed on the target capture part in the video image to obtain the target recognition result; thus The virtual camera performs image rendering based on the target recognition result to obtain a virtual live image, and then determines the video stream data according to the multiple virtual live images.

Afterwards, the video stream data can be transmitted to the mobile terminal device through the CDN distribution network, and the video stream data can be transmitted to the PC device through the CDN distribution network and the forwarding stream server, so as to play the video stream data in the viewer device.

In the embodiment of the present invention, the method of determining the target capture position through the virtual live broadcast mode and performing action recognition on the target capture position can achieve a more flexible determination of the body parts to be captured of the real anchor. Through this processing method, it is possible to filter out There is no need for body parts for motion capture, thereby shortening the time for motion capture and reducing the complexity of the virtual live broadcast. At the same time, it can also make the live broadcast method require live broadcast scenes with high immediacy.

Those skilled in the art can understand that in the above method of specific implementation, the writing order of each step does not mean a strict execution order and constitutes any limitation on the implementation process. The specific execution order of each step should be based on its function and possible The inner logic is OK.

Based on the same inventive concept, the embodiment of the present invention also provides a live broadcast device corresponding to the live broadcast method. Since the problem-solving principle of the device in the embodiment of the present invention is similar to the above-mentioned live broadcast method in the embodiment of the present invention, the implementation of the device can refer to the method The implementation of this method will not be repeated here.

Referring to FIG. 9 , it is a schematic diagram of a live broadcast device provided by an embodiment of the present invention, and the device includes: a first determination unit 81, an acquisition unit 82, an action recognition unit 83, and a second determination unit 84; wherein, The first determining unit 81 is configured to determine a virtual live broadcast mode; the virtual live broadcast mode is used to indicate the target capture position for body capture of the real anchor; The collection unit 82 is configured to collect the video images of the real anchor during the live broadcast; The motion recognition unit 83 is configured to perform motion recognition on the target capture part in the video image based on the virtual live broadcast mode, and obtain a target recognition result; The second determining unit 84 is configured to determine the video stream data corresponding to the virtual anchor model driven by the real anchor based on the target recognition result; the video stream data is used to present the result of the virtual anchor model executing the recognition result Indicates the course of action.

In the embodiment of the present invention, after the virtual live broadcast mode is determined, the virtual live broadcast mode determines the target capture position, without manually setting the target capture position, and can automatically filter out the body parts that do not need motion capture, thereby realizing automatic The body parts to be captured of the real anchor are captured, thereby shortening the time for motion capture and reducing the complexity of the virtual live broadcast. At the same time, the live broadcast method can also be adapted to live broadcast scenes with high immediacy.

In a possible implementation manner, the first determination unit 81 is further configured to: determine the running scenario of the anchor device; The target running version; determine the virtual live broadcast mode based on the determined target running version.

In a possible implementation manner, the first determination unit 81 is further configured to: acquire the device identification information of the host device in the case of detecting the first installation operation for the live application program, and according to the device identification information Determining the running scenario; and/or, in the case of detecting the opening operation for the live broadcast application, obtaining the remaining device computing resources of the anchor device at the current moment, and determining the remaining device computing resources according to the remaining device computing resources Describe the running scenario.

In a possible implementation manner, the first determining unit 81 is further configured to: respond to a selection instruction of the live broadcast mode of the real anchor, and determine the virtual live broadcast mode among the plurality of preset live broadcast modes.

In a possible implementation manner, the device is further configured to: collect a preview image containing the real anchor before collecting the video image of the real anchor during the live broadcast; determine the Whether the target capture part satisfies the motion recognition condition; if the target capture part contained in the preview image does not meet the motion recognition condition, generate target adjustment information until it is determined that the target capture part meets the motion recognition Conditions, wherein the target adjustment information is used to remind the real anchor to adjust the display state of the target capture part in the preview image.

In a possible implementation manner, the action recognition unit 83 is further configured to: acquire the mode tag corresponding to the virtual live broadcast mode when it is detected that the target capture part includes a hand part, wherein the mode tag A target pattern tag including whether gesture recognition is performed on the hand part; if the pattern tag is determined to be the target pattern tag, perform motion recognition on the target capture part in the video image, and perform action recognition on the target capture part Gesture detection is performed on the hand parts in the object, and the target recognition result including the gesture recognition result is obtained.

In a possible implementation manner, the action recognition unit 83 is further configured to: detect that the hand gesture of the real anchor is a preset gesture according to the hand detection result, and acquire the rendering material corresponding to the preset gesture special effects; rendering the rendering material special effects in the specified video frame in the video stream data.

In a possible implementation manner, the second determination unit 84 is further configured to: acquire a live viewing request sent by at least one viewer device; determine the interface background image of the live viewing interface corresponding to each viewer device based on the live viewing request image, wherein the interface background image includes: a static background image or a dynamic background image; a plurality of video images determined to characterize the virtual anchor model performing the action indicated by the target recognition result; The background image of each video image in the plurality of video images is replaced with the interface background image, and the video stream data is determined based on the modified video images.

In a possible implementation manner, the device is further configured to: after determining the video stream data corresponding to the virtual anchor model driven by the real anchor, push a video containing the interface background image to the audience device corresponding to the same interface background image. video stream data.

In a possible implementation manner, the device is further configured to: display the first indication information and/or the second indication information in the display interface of the anchor device after determining the virtual live broadcast mode; wherein, the first indication information The indication information is used to indicate the target capture part in the valid capture state, and the second indication information is used to indicate the target capture part in the invalid capture state.

For the description of the processing flow of each module in the device and the interaction flow between the modules, reference may be made to the relevant description in the above method embodiment, and details will not be described here.

Corresponding to the live broadcast method in FIG. 1A, the embodiment of the present invention also provides a computer device 900, as shown in FIG. 10, which is a schematic structural diagram of the electronic device 900 provided by the embodiment of the present invention, including: Processor 91, memory 92, and bus 93; memory 92 is used to store execution instructions, including internal memory 921 and external memory 922; internal memory 921 is used to temporarily store computing data in the processor 91, and For the data exchanged with the external memory 922 such as hard disk, the processor 91 exchanges data with the external memory 922 through the internal memory 921. When the electronic device 900 is running, the connection between the processor 91 and the memory 92 Inter-communication through the bus 93, so that the processor 91 executes the following instructions: Determine the virtual live broadcast mode; the virtual live broadcast mode is used to indicate the target capture site for body capture of the real anchor; Collecting the video images of the real anchor during the live broadcast; Based on the virtual live broadcast mode, perform action recognition on the target capture part in the video image to obtain a target recognition result; Based on the target recognition result, determine the video stream material corresponding to the virtual anchor model driven by the real anchor; the video stream material is used to present the process of the virtual anchor model executing the action indicated by the recognition result.

An embodiment of the present invention also provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the steps of the live broadcasting method described in the foregoing method embodiments are executed. Wherein, the storage medium may be a volatile or non-volatile computer-readable storage medium.

The embodiment of the present invention also provides a computer program product, the computer program product carries a program code, and the instructions contained in the program code can be used to execute the steps of the live broadcast method described in the above method embodiment, please refer to the above method embodiment, I won't repeat them here.

Wherein, the above-mentioned computer program product can be realized by means of hardware, software or a combination thereof. In an optional embodiment, the computer program product is embodied as a computer storage medium. In another optional embodiment, the computer program product is embodied as a software product, such as a software development kit and the like.

The device involved in the embodiments of the present invention may be at least one of a system, a method, and a computer program product. The computer program product may include a computer-readable storage medium loaded with computer-readable program instructions configured to cause a processor to implement various aspects of the invention.

A computer readable storage medium may be a tangible device that can hold and store instructions for use by an instruction execution device. The computer-readable storage medium may be, for example but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. Examples (non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, Random Access Memory (RAM), Read-Only Memory (ROM) , Erasable Programmable Read Only Memory (Electrical Programmable Read Only Memory, EPROM) or flash memory, Static Random-Access Memory (Static Random-Access Memory, SRAM), portable compressed disk read-only Compact Disc Read-Only Memory (CD-ROM), Digital Video Disc (DVD), memory sticks, floppy disks, mechanically encoded devices such as punched cards or slots on which instructions are stored Inner protrusion structure, and any suitable combination of the above. As used herein, computer-readable storage media are not to be construed as transient signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other Electrical signals transmitted by wires.

The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing/processing device, or to an external computer over at least one of a network, such as the Internet, a local area network, a wide area network, and a wireless network or external storage device. The network may include at least one of copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and edge servers. The network interface card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in computer-readable storage in each computing/processing device in the media.

Computer program instructions configured to perform the operations of the present invention may be assembly instructions, Instruction Set Architecture (Industry Standard Architecture, ISA) instructions, machine instructions, machine-related instructions, microcode, firmware instructions, state setting data, or Source code or object code written in any combination of design languages, including object-oriented programming languages—such as Smalltalk, C++, etc., and conventional procedural programming languages, such as the "C" language or similar programming language. Computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone package, partly on the user's computer and partly on a remote computer, or entirely on a remote computer. Execute on the terminal computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any kind of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or it may be connected to An external computer (e.g. using an Internet service provider to connect via the Internet). In some embodiments, electronic circuits, such as programmable logic circuits, FPGAs, or programmable logic arrays (Programmable Logic Arrays, PLAs), can be customized by utilizing state information of computer-readable program instructions, which can execute Computer-readable program instructions to realize various aspects of the embodiments of the present invention.

Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process of the above-described system and device can refer to the corresponding process in the foregoing method embodiments, which will not be repeated here. In the several embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or elements can be combined or can be Integrate into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some communication interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or may also be distributed to multiple network units . Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.

If the functions described above are realized in the form of software function units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium executable by a processor. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of software products, which are stored in a storage medium, including several The instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk, etc., which can store program codes. media.

Finally, it should be noted that: the above-described embodiments are only specific implementations of the present invention, used to illustrate the technical solutions of the present invention, rather than limiting them, and the scope of protection of the present invention is not limited thereto, although referring to the foregoing The embodiment has described the present invention in detail, and those skilled in the art should understand that any person familiar with the technical field can still modify the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention Changes can be easily thought of, or equivalent replacements are made to some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of the present invention within the scope of protection. Therefore, the protection scope of the present invention should be based on the protection scope of the patent application.

Industrial Applicability An embodiment of the present invention provides a live broadcast method, system, computer equipment, and computer-readable storage medium, wherein the method includes: determining a virtual live broadcast mode; the virtual live broadcast mode is used to indicate at least one target for body capture of a real anchor Capture the location; collect the video image of the real anchor during the live broadcast; based on the virtual live broadcast mode, perform action recognition on the target capture location in the video image to obtain a target recognition result; based on the target recognition result , determining the video stream material corresponding to the virtual anchor model driven by the real anchor; the video stream material is used to present a process in which the virtual anchor model performs the action indicated by the recognition result.

201: video image acquisition terminal 202: Network 203: Control terminal 71:Anchor terminal equipment 72: Audience device 81: The first determination unit 82: Acquisition unit 83: Action recognition unit 84: The second determination unit 900: computer equipment 91: Processor 92: Memory 921: internal memory 922: external memory 93: busbar S101~S107, S11~S13: steps

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the accompanying drawings used in the embodiments will be briefly introduced below, and the drawings here are incorporated into the specification and constitute a part of the specification. The drawings show the embodiments consistent with the present invention, and are used together with the description to illustrate the technical solution of the present invention. It should be understood that the following drawings only show some embodiments of the present invention, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, they can also make From these figures are obtained other related figures. FIG. 1A shows a flow chart of a live broadcast method provided by an embodiment of the present invention; FIG. 1B shows a schematic diagram of a system architecture to which the live broadcast method according to the embodiment of the present invention can be applied; FIG. 2 shows a flow chart of another live broadcast method provided by an embodiment of the present invention; FIG. 3 shows a schematic interface diagram of a selection interface of a virtual live broadcast mode provided by an embodiment of the present invention; FIG. 4 shows a schematic interface diagram of another virtual live mode selection interface provided by an embodiment of the present invention; FIG. 5 shows a schematic interface diagram of a selection interface of a third virtual live broadcast mode provided by an embodiment of the present invention; FIG. 6 shows a schematic interface diagram of a mode label setting interface provided by an embodiment of the present invention; FIG. 7 shows a schematic interface diagram of a virtual live broadcast interface including gesture information of preset gestures provided by an embodiment of the present invention; Fig. 8 shows a schematic diagram of a live broadcast system provided by an embodiment of the present invention; Fig. 9 shows a schematic diagram of a live broadcast device provided by an embodiment of the present invention; Fig. 10 shows a schematic diagram of a computer device provided by an embodiment of the present invention.

S101~S107: steps

Claims (14)

A live broadcast method, applied to anchor equipment, including: Determine the virtual live broadcast mode; the virtual live broadcast mode is used to indicate at least one target capture site for body capture of the real anchor; Collecting the video images of the real anchor during the live broadcast; Based on the virtual live broadcast mode, perform action recognition on the target capture part in the video image to obtain a target recognition result; Based on the target recognition result, determine the video stream material corresponding to the virtual anchor model driven by the real anchor; the video stream material is used to present a process in which the virtual anchor model executes the action indicated by the target recognition result. According to the method described in claim 1, wherein said determining the virtual live broadcast mode includes: Determine the operating scenario of the anchor device; Determining a target running version matching the running scenario based on multiple running versions pre-set for the anchor device; The virtual live broadcast mode is determined based on the determined target running version. According to the method described in claim 2, wherein the determining the operation scenario of the anchor device includes: In the case of detecting the first installation operation for the live broadcast application, obtaining the device identification information of the anchor device, and determining the running scenario according to the device identification information; and/or When an opening operation for the live broadcast application is detected, the remaining device computing resources of the host device at the current moment are acquired, and the running scenario is determined according to the remaining device computing resources. The method according to any one of claims 1 to 3, wherein said determining the virtual live broadcast mode includes: In response to the live broadcast mode selection instruction of the real host, the virtual live broadcast mode is determined among a plurality of preset live broadcast modes. The method according to any one of claims 1 to 3, wherein the method further comprises: Before collecting the video images of the real anchor during the live broadcast, collecting a preview image that includes the real anchor; determining whether the target capture part contained in the preview image satisfies the motion recognition condition; If the target capture part included in the preview image does not satisfy the motion recognition condition, generate target adjustment information until it is determined that the target capture part satisfies the motion recognition condition, wherein the target adjustment information It is used to remind the real anchor to adjust the display state of the target capturing part in the preview image. According to the method described in claim 1, wherein, based on the virtual live broadcast mode, the action recognition is performed on the target capture part in the video image, and the target recognition result is obtained, including: When it is detected that the target capture part includes a hand part, the mode tag corresponding to the virtual live broadcast mode is obtained, wherein the mode tag includes a target mode tag whether to perform hand recognition on the hand part; When it is determined that the pattern tag is the target pattern tag, perform action recognition on the target capture part in the video image, and perform hand detection on the hand parts in the target capture part, and obtain The target recognition result of the partial detection result. The method according to claim 6, wherein, after determining the video stream data corresponding to the virtual anchor model driven by the real anchor, the method further includes: According to the hand detection result, it is detected that the hand gesture of the real anchor is a preset gesture, and the rendering material special effect corresponding to the preset gesture is obtained; Rendering the rendering material special effect in a specified video frame in the video stream data. The method according to any one of claims 1 to 3, wherein, based on the target recognition result, determining the video stream data corresponding to the virtual anchor model driven by the real anchor includes: Obtain a live viewing request sent by at least one viewer device; Determine the interface background image of the live viewing interface corresponding to each audience device based on the live viewing request, wherein the interface background image includes: a static background image or a dynamic background image; Determining a plurality of video images used to characterize the virtual anchor model performing the action indicated by the target recognition result; The background image of each video image in the plurality of video images is replaced with the interface background image, and the video stream data is determined based on the modified video images. The method according to claim 8, wherein the method further comprises: After the video stream data corresponding to the virtual anchor model driven by the real anchor is determined, the video stream data including the interface background image is pushed to the audience device corresponding to the same interface background image. The method according to claim 1, wherein the method further comprises: After the virtual live broadcast mode is determined, display the first indication information and/or the second indication information in the display interface of the anchor device; wherein, the first indication information is used to indicate the target capture position in an effective capture state, The second indication information is used to indicate target capture parts in an invalid capture state. A live broadcast system, comprising: a host device and a viewer device; The anchor device is configured to determine the video stream data corresponding to the virtual anchor model driven by the real anchor according to the live broadcast method described in any one of the above request items 1 to 10, and push the video stream data to the viewer end device. the video streaming data; The viewer device is configured to obtain the video stream data, and play the video stream data on a live viewing interface. The system according to claim 11, wherein the viewer device includes: a mobile terminal device and a PC device; When the viewer device is the mobile terminal device, the anchor device transmits the video stream data to the viewer device through a CDN distribution network; In the case that the viewer-end device is the PC device, the host-end device transmits the video stream data to the viewer-end device through a CDN distribution network and a forwarding stream server. A computer device, comprising: a processor, a memory, and a bus bar, the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the connection between the processor and the memory Through bus communication, when the machine-readable instructions are executed by the processor, the steps of the live broadcasting method as described in any one of claim items 1 to 10 are performed. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the steps of the live broadcast method as described in any one of claims 1 to 10 are executed.

Images