Digital human system application method and apparatus

Abstract

The present application provides a digital human system application method and apparatus. The method specifically comprises: connecting a display device, and starting a pre-stored digital human system by means of a startup script; acquiring three-dimensional model data and human body action data of a digital human in the digital human system; decomposing the three-dimensional model data and the human body action data into a plurality of processing tasks by means of a rendering engine, and transmitting the plurality of processing tasks to a preset graphics processing unit; and acquiring a digital human picture generated by the graphics processing unit on the basis of the plurality of processing tasks, and sending the digital human picture to the display device. The present application facilitates improvement of the application efficiency of the digital human system.

Description

A method and apparatus for applying a digital human system

[0001] Cross-referencing related documents

[0002] This application claims priority to Chinese Patent Application No. 2024115938121, filed on November 8, 2024, entitled "A Method and Apparatus for the Application of a Digital Human System", the entire contents of which are incorporated herein by reference. Technical Field

[0003] The embodiments of this application belong to the field of Internet technology, and in particular relate to a method and apparatus for applying a digital human system. Background Technology

[0004] Digital human systems are virtual character systems built using deep learning and advanced algorithms. These systems possess highly realistic appearances, voices, and movements, and can interact using natural language, understand complex situations, and react accordingly, just like humans.

[0005] However, existing digital human systems often rely heavily on large computing devices, especially servers equipped with high-performance graphics cards. While servers can provide powerful data processing and graphics rendering capabilities, their inherently large size and complex deployment environment limit the application scenarios and flexibility of digital human systems. This is because complex deployment environments often significantly increase the manual setup and deployment steps for users. Therefore, how to simplify the application process of digital human systems is an urgent problem to be solved. Technical solutions

[0006] This application provides a method and apparatus for applying a digital human system, in order to solve the aforementioned technical problem of how to simplify the application process of a digital human system.

[0007] In a first aspect, embodiments of this application provide a method for applying a digital human system to a portable device, the method comprising:

[0008] Connect to the display device and start the pre-stored digital human system via the startup script;

[0009] Acquire the 3D model data and human motion data of the digital human in the digital human system;

[0010] The rendering engine decomposes the 3D model data and the human motion data into multiple processing tasks, and transmits the multiple processing tasks to a preset graphics processor.

[0011] The graphics processor acquires the digital human image generated based on multiple processing tasks and sends the digital human image to the display device.

[0012] The beneficial effects of this application embodiment are twofold. Firstly, by connecting to a display device and starting a pre-stored digital human system via a startup script, the pre-stored digital human system is launched immediately upon connection to the display device, thus achieving "connect and use immediately," which improves the speed of using the digital human system. Secondly, by acquiring the 3D model data and human motion data of the digital human in the digital human system, and decomposing the 3D model data and human motion data into multiple processing tasks through a rendering engine, transmitting these multiple processing tasks to a preset graphics processor, acquiring the digital human image generated by the graphics processor based on the multiple processing tasks, and sending the digital human image to the display device, the manual setup and deployment steps for the user are reduced, thus simplifying the application process of the digital human system and improving its application efficiency.

[0013] In one possible implementation of the first aspect, a display device is connected, and a pre-stored digital human system is started via a startup script, including:

[0014] The hot-plug signal sent by the display device is obtained through the input / output interface, which includes one or a combination of HDMI interface, USB interface and Ethernet interface;

[0015] When the hot-plug signal is a high-level signal, a data transmission channel is established between the portable device and the display device;

[0016] When the data transmission channel meets the preset conditions, the pre-stored digital human system is started by starting the startup script.

[0017] In this embodiment, a data transmission channel is established between the portable device and the display device to ensure that the portable device can stably transmit digital human images to the display device, thereby improving the efficiency and stability of digital human image transmission.

[0018] In one possible implementation of the first aspect, the step of activating the pre-stored digital human system via a startup script when the data transmission channel meets preset conditions includes:

[0019] The transmission delay duration of the data transmission channel is obtained. When the transmission delay duration is less than the preset delay duration in the preset conditions, the pre-stored digital human system is started by starting the startup script.

[0020] Alternatively, the transmission rate of the data transmission channel can be obtained, and when the transmission rate is greater than the preset rate in the preset conditions, the pre-stored digital human system can be started by starting the startup script.

[0021] In this embodiment, when the data transmission channel meets preset conditions, a pre-stored digital human system is launched via a startup script. The digital human system can automatically avoid network congestion or unstable periods, effectively reducing adverse situations such as startup failures, data transmission interruptions, or delays caused by network problems, thus enhancing the stability and reliability of the digital human system. From a user experience perspective, a transmission delay shorter than a preset delay or a transmission rate greater than a preset rate provides users with a smoother and faster data interaction experience, enhancing user satisfaction and loyalty.

[0022] In one possible implementation of the first aspect, acquiring the 3D model data and human motion data of the digital human in the digital human system includes:

[0023] Obtain the first and second files of the digital human system;

[0024] In the first file, the three-dimensional model data of the digital human is obtained, and in the second file, the human motion data of the digital human is obtained.

[0025] In this embodiment of the application, a digital human can be reconstructed using 3D model data and human motion data, ensuring that the digital human's motion performance is completely consistent with the digital human system. This real-time feedback mechanism not only improves the user experience but also enhances the interactivity and immersion of the digital human.

[0026] In one possible implementation of the first aspect, the step of decomposing the 3D model data and the human motion data into multiple processing tasks using a rendering engine, and transmitting the multiple processing tasks to a preset graphics processor, includes:

[0027] The 3D model data and the human motion data are combined to form rendering data;

[0028] The rendering engine decomposes the rendering task of the rendering data into multiple processing tasks, and transmits the multiple processing tasks to a preset graphics processor.

[0029] In this embodiment, the graphics processing unit (GPU) possesses powerful parallel computing capabilities, containing a large number of processing units and multiple stream processors. When multiple processing tasks are simultaneously sent to the GPU, these tasks can be executed in parallel on multiple processing units of the GPU. This parallel processing method significantly improves computational efficiency, enabling tasks that would otherwise take a long time to complete to be processed in a shorter time, thereby accelerating the overall workflow.

[0030] In one possible implementation of the first aspect, after acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device, the digital human system application method further includes:

[0031] Collect user voice and convert the user voice into first text information;

[0032] The first text information is input into a deep learning model to obtain the second text information output by the deep learning model based on the first text information.

[0033] The second text information is converted into a response voice, and the response voice is sent to the display device.

[0034] In this embodiment of the application, in the digital age, users expect their interaction with display devices to be more natural and smooth. By having the display device broadcast response voice, it not only responds to the user's voice but also provides feedback in a more humanized way, enhancing the user's sense of participation and immersion.

[0035] In one possible implementation of the first aspect, the first text information is input into a deep learning model to obtain second text information output by the deep learning model based on the first text information, including:

[0036] Obtain the model file of the deep learning model, and load the deep learning model from the model file using a loading script;

[0037] The first text information is input into the deep learning model to obtain the second text information output by the deep learning model based on the first text information. In this embodiment, the model file serves as the carrier of the deep learning model, containing the network parameters and structural information optimized during the training process. These parameters and structural information are the foundation for the deep learning model to perform prediction and inference. By loading the model file, the deep learning model can quickly restore its complete state without having to undergo a time-consuming retraining process, thus improving the convenience and efficiency of deep learning model applications.

[0038] In one possible implementation of the first aspect, after acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device, the digital human system application method further includes:

[0039] Obtain an update command, retrieve update content from the server using the update command, perform an update operation on the digital human system using the update content, and obtain the updated digital human system.

[0040] In the embodiments of this application, the updated digital human system typically possesses enhanced interactive capabilities. This includes more precise natural language processing technology, enabling the digital human to better understand the user's intentions and emotions, and provide more appropriate and human-like responses.

[0041] In one possible implementation of the first aspect, after acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device, the digital human system application method further includes:

[0042] Obtain a push command, execute the push command, and push the digital human image to a preset database.

[0043] In this embodiment, the push command is executed to push the digital human image to a preset database. The database, as a persistent storage medium, can permanently store the digital human image on the disk, ensuring that the image is not lost even in the event of a system power outage or crash. This persistent storage mechanism provides strong protection for the long-term preservation and reliability of the digital human image.

[0044] Secondly, embodiments of this application provide a digital human system application device, including:

[0045] The connection module is used to connect to the display device and start the pre-stored digital human system via a startup script;

[0046] The acquisition module is used to acquire the 3D model data and human motion data of the digital human in the digital human system;

[0047] The transmission module is used to decompose the 3D model data and the human motion data into multiple processing tasks through the rendering engine, and transmit the multiple processing tasks to a preset graphics processor.

[0048] The sending module is used to acquire the digital human image generated by the graphics processor based on multiple processing tasks, and send the digital human image to the display device. Beneficial effects

[0049] The beneficial effects of this application embodiment are twofold. Firstly, by connecting to a display device and starting a pre-stored digital human system via a startup script, the pre-stored digital human system is launched immediately upon connection to the display device, thus achieving "connect and use immediately," which improves the speed of using the digital human system. Secondly, by acquiring the 3D model data and human motion data of the digital human in the digital human system, and decomposing the 3D model data and human motion data into multiple processing tasks through a rendering engine, transmitting these multiple processing tasks to a preset graphics processor, acquiring the digital human image generated by the graphics processor based on the multiple processing tasks, and sending the digital human image to the display device, the manual setup and deployment steps for the user are reduced, thus simplifying the application process of the digital human system and improving its application efficiency. Attached Figure Description

[0050] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. Some specific embodiments of this application will be described in detail below with reference to the accompanying drawings in an exemplary and non-limiting manner. The same reference numerals in the drawings designate the same or similar parts or components. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings:

[0051] Figure 1 is a flowchart of the digital human system application method provided in an embodiment of this application;

[0052] Figure 2 is a schematic block diagram of a digital human system application device provided in an embodiment of this application. Detailed Implementation

[0053] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some, not all, of the embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative effort should fall within the scope of protection of the present application.

[0054] The digital human system application method provided in this application embodiment can be applied to portable devices, which are small, lightweight, and easy-to-carry and move electronic devices.

[0055] The portable device is equipped with a common deep learning model, inference chip, graphics processor, storage device, data I / O interface, and power interface.

[0056] The deep learning model of the digital human is deployed on the storage device, and the inference chip completes the inference operation of the deep learning model.

[0057] To ensure that the inference chip supports the operators contained in the trained deep learning model, the inference chip needs to adopt a general-purpose CPU architecture and a GPU architecture.

[0058] The I / O interfaces include, but are not limited to, HDMI interfaces, USB interfaces, and Ethernet interfaces.

[0059] In this context, the Chinese translation of IO interface is "input / output interface".

[0060] The HDMI interface, short for High Definition Multimedia Interface, is a digital video / audio interface that can transmit audio and video signals simultaneously without requiring digital-to-analog or analog-to-digital conversion before signal transmission.

[0061] USB, short for Universal Serial Bus, is a widely used interface standard for connecting external devices such as keyboards, mice, printers, and storage devices.

[0062] Ethernet interface, also known as Ethernet interface, is a network data connection interface and is one of the most widely used local area network communication methods.

[0063] The operating system of portable devices adds mounting scripts for USB and HDMI interfaces. These mounting scripts are used to detect whether the software deployed on the box contains a startup script.

[0064] The startup script is a code file containing the content required to execute and load the pre-stored digital human system.

[0065] The display of digital humans relies on graphics processing units (GPUs) to provide high-quality graphics rendering capabilities. GPUs have significant advantages in graphics rendering, image processing, and modeling, and can process complex graphics data in real time.

[0066] In the field of travel guides, users can obtain local tour information and voice navigation services through the digital human system on portable devices;

[0067] In the field of personal assistants, users can carry a portable digital human system to interact with others via voice and obtain personalized services such as schedule reminders and weather forecasts.

[0068] Figure 1 is a flowchart of the digital human system application method provided in an embodiment of this application. As shown in Figure 1, the digital human system application method provided in this embodiment of this application is applied to the aforementioned portable device. The digital human system application method includes the following steps, which are detailed below:

[0069] S101 connects to a display device and starts a pre-stored digital human system via a startup script;

[0070] In one possible implementation of the first aspect, a display device is connected, and a pre-stored digital human system is started via a startup script, including:

[0071] The hot-plug signal sent by the display device is obtained through the input / output interface, which includes one or a combination of HDMI interface, USB interface and Ethernet interface;

[0072] When the hot-plug signal is a high-level signal, a data transmission channel is established between the portable device and the display device;

[0073] When the data transmission channel meets the preset conditions, the pre-stored digital human system is started by starting the startup script.

[0074] In this embodiment, a data transmission channel is established between the portable device and the display device to ensure that the portable device can stably transmit digital human images to the display device, thereby improving the efficiency and stability of digital human image transmission.

[0075] In one possible implementation of the first aspect, the step of activating the pre-stored digital human system via a startup script when the data transmission channel meets preset conditions includes:

[0076] The transmission delay duration of the data transmission channel is obtained. When the transmission delay duration is less than the preset delay duration in the preset conditions, the pre-stored digital human system is started by starting the startup script.

[0077] Alternatively, the transmission rate of the data transmission channel can be obtained, and when the transmission rate is greater than the preset rate in the preset conditions, the pre-stored digital human system can be started by starting the startup script.

[0078] In this embodiment, when the data transmission channel meets preset conditions, a pre-stored digital human system is launched via a startup script. The digital human system can automatically avoid network congestion or unstable periods, effectively reducing adverse situations such as startup failures, data transmission interruptions, or delays caused by network problems, thus enhancing the stability and reliability of the digital human system. From a user experience perspective, a transmission delay shorter than a preset delay or a transmission rate greater than a preset rate provides users with a smoother and faster data interaction experience, enhancing user satisfaction and loyalty.

[0079] S102, acquire the 3D model data and human motion data of the digital human in the digital human system;

[0080] The acquisition of the 3D model data and human motion data of the digital human in the digital human system includes:

[0081] Obtain the first and second files of the digital human system;

[0082] In the first file, the three-dimensional model data of the digital human is obtained, and in the second file, the human motion data of the digital human is obtained.

[0083] In this embodiment of the application, a digital human can be reconstructed using 3D model data and human motion data, ensuring that the digital human's motion performance is completely consistent with the digital human system. This real-time feedback mechanism not only improves the user experience but also enhances the interactivity and immersion of the digital human.

[0084] S103, the rendering engine decomposes the three-dimensional model data and the human motion data into multiple processing tasks, and transmits the multiple processing tasks to a preset graphics processor;

[0085] The step of decomposing the 3D model data and the human motion data into multiple processing tasks through a rendering engine, and transmitting the multiple processing tasks to a preset graphics processor, includes:

[0086] The 3D model data and the human motion data are combined to form rendering data;

[0087] The rendering engine decomposes the rendering task of the rendering data into multiple processing tasks, and transmits the multiple processing tasks to a preset graphics processor.

[0088] In this embodiment, the graphics processing unit (GPU) possesses powerful parallel computing capabilities, containing a large number of processing units and multiple stream processors. When multiple processing tasks are simultaneously sent to the GPU, these tasks can be executed in parallel on multiple processing units of the GPU. This parallel processing method significantly improves computational efficiency, enabling tasks that would otherwise take a long time to complete to be processed in a shorter time, thereby accelerating the overall workflow.

[0089] S104, acquire the digital human image generated by the graphics processor based on multiple processing tasks, and send the digital human image to the display device.

[0090] For example, acquiring a digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device includes:

[0091] The graphics processor acquires the digital human image generated based on multiple processing tasks, decomposes the content of the digital human image into image frames, and sends the image frames to the display device.

[0092] Once the display device receives the digital human image, it can smoothly present the digital human's appearance, movements, and expressions to the user.

[0093] Portable devices can be connected directly to display devices or indirectly to display devices.

[0094] For ease of explanation, the following example is provided:

[0095] For example, when only a display device is available, the portable device can be directly connected to it. In this application, the portable device's HDMI port is connected to the display device's HDMI port, its Ethernet port to the network, and its power port to the power source. The HDMI mounting script detects the startup script and starts the digital human system. The portable device then transmits the digital human image from the digital human system to the display device via the HDMI port, and the display device shows the digital human image.

[0096] For example, if a computer is available, the portable device connects to the computer, and the computer connects to the display device. Therefore, the portable device is indirectly connected to the display device. In application, the portable device's USB port is connected to the computer, its Ethernet port is connected to the network, and its power port is connected to a power source. The USB port's mounting script detects the startup script and starts the digital human system.

[0097] The portable device transmits the digital human image from the digital human system to the computer via a USB interface. The computer then transmits the digital human image from the digital human system to a display device, which displays the digital human image from the digital human system.

[0098] The digital human system application method further includes, after acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device:

[0099] Step A: Collect user voice and convert the user voice into first text information;

[0100] The first text information is input into a deep learning model to obtain the second text information output by the deep learning model based on the first text information.

[0101] The second text information is converted into a response voice, and the response voice is sent to the display device.

[0102] For example, converting the second text information into a response voice and sending the response voice to the display device includes:

[0103] The second text information is converted into a response voice, and the second text information and the response voice are sent to the display device so that the display device displays the second text information and plays the response voice.

[0104] In this embodiment of the application, in the digital age, users expect their interaction with display devices to be more natural and smooth. By having the display device broadcast response voice, it not only responds to the user's voice but also provides feedback in a more humanized way, enhancing the user's sense of participation and immersion.

[0105] The process of inputting the first text information into a deep learning model and obtaining the second text information output by the deep learning model based on the first text information includes:

[0106] Obtain the model file of the deep learning model, and load the deep learning model from the model file using a loading script;

[0107] The first text information is input into the deep learning model to obtain the second text information output by the deep learning model based on the first text information. In this embodiment, the model file serves as the carrier of the deep learning model, containing the network parameters and structural information optimized during the training process. These parameters and structural information are the foundation for the deep learning model to perform prediction and inference. By loading the model file, the deep learning model can quickly restore its complete state without having to undergo a time-consuming retraining process, thus improving the convenience and efficiency of deep learning model applications.

[0108] The digital human system application method further includes, after acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device:

[0109] Step B: Obtain the update command, retrieve the update content from the server using the update command, and perform an update operation on the digital human system using the update content to obtain the updated digital human system.

[0110] In the embodiments of this application, the updated digital human system typically possesses enhanced interactive capabilities. This includes more precise natural language processing technology, enabling the digital human to better understand the user's intentions and emotions, and provide more appropriate and human-like responses.

[0111] The digital human system application method further includes, after acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device:

[0112] Step C: Obtain a push command, execute the push command, and push the digital human image to a preset database.

[0113] In this embodiment, the push command is executed to push the digital human image to a preset database. The database, as a persistent storage medium, can permanently store the digital human image on the disk, ensuring that the image is not lost even in the event of a system power outage or crash. This persistent storage mechanism provides strong protection for the long-term preservation and reliability of the digital human image.

[0114] Steps A, B, and C can be executed simultaneously, or they can be executed at different times.

[0115] For example, step A can be executed before or after steps B and C, step B can be executed before or after steps A and C, and step C can be executed before or after steps A and B. The specific execution order is not restricted here.

[0116] The beneficial effects of this application embodiment are twofold. Firstly, by connecting to a display device and starting a pre-stored digital human system via a startup script, the pre-stored digital human system is launched immediately upon connection to the display device, thus achieving "connect and use immediately," which improves the speed of using the digital human system. Secondly, by acquiring the 3D model data and human motion data of the digital human in the digital human system, and decomposing the 3D model data and human motion data into multiple processing tasks through a rendering engine, transmitting these multiple processing tasks to a preset graphics processor, acquiring the digital human image generated by the graphics processor based on the multiple processing tasks, and sending the digital human image to the display device, the manual setup and deployment steps for the user are reduced, thus simplifying the application process of the digital human system and improving its application efficiency.

[0117] Corresponding to the digital human system application method described in the above embodiments, please refer to Figure 2. Figure 2 is a schematic block diagram of the digital human system application device provided in the embodiments of this application. The digital human system application device 200 shown in Figure 2 can be applied to the above-mentioned electronic devices. The digital human system application device 200 shown in Figure 2 will be described in detail below using an electronic device as an example. The digital human system application device 200 may include a connection module 201, an acquisition module 202, a transmission module 203, and a sending module 204.

[0118] The connection module 201 is used to connect to the display device and start the pre-stored digital human system through the startup script;

[0119] The acquisition module 202 is used to acquire the three-dimensional model data and human motion data of the digital human in the digital human system;

[0120] The transmission module 203 is used to decompose the three-dimensional model data and the human motion data into multiple processing tasks through the rendering engine, and transmit the multiple processing tasks to a preset graphics processor.

[0121] The sending module 204 is used to acquire the digital human image generated by the graphics processor based on multiple processing tasks, and send the digital human image to the display device.

[0122] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0123] The beneficial effects of this application embodiment are twofold. Firstly, by connecting to a display device and starting a pre-stored digital human system via a startup script, the pre-stored digital human system is launched immediately upon connection to the display device, thus achieving "connect and use immediately," which improves the speed of using the digital human system. Secondly, by acquiring the 3D model data and human motion data of the digital human in the digital human system, and decomposing the 3D model data and human motion data into multiple processing tasks through a rendering engine, transmitting these multiple processing tasks to a preset graphics processor, acquiring the digital human image generated by the graphics processor based on the multiple processing tasks, and sending the digital human image to the display device, the manual setup and deployment steps for the user are reduced, thus simplifying the application process of the digital human system and improving its application efficiency.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for applying a digital human system, characterized in that, The application method of the digital human system includes: Connect to the display device and start the pre-stored digital human system via the startup script; Acquire the 3D model data and human motion data of the digital human in the digital human system; The rendering engine decomposes the 3D model data and the human motion data into multiple processing tasks, and transmits the multiple processing tasks to a preset graphics processor. The graphics processor acquires the digital human image generated based on multiple processing tasks and sends the digital human image to the display device.

2. The application method of the digital human system according to claim 1, characterized in that, Connect to a display device and, via a startup script, launch a pre-stored digital human system, including: The hot-plug signal sent by the display device is obtained through the input / output interface, which includes one or a combination of HDMI interface, USB interface and Ethernet interface; When the hot-plug signal is a high-level signal, a data transmission channel is established between the portable device and the display device; When the data transmission channel meets the preset conditions, the pre-stored digital human system is started by starting the startup script.

3. The application method of the digital human system according to claim 2, characterized in that, When the data transmission channel meets preset conditions, the pre-stored digital human system is started via a startup script, including: The transmission delay duration of the data transmission channel is obtained. When the transmission delay duration is less than the preset delay duration in the preset conditions, the pre-stored digital human system is started by starting the startup script. Alternatively, the transmission rate of the data transmission channel can be obtained, and when the transmission rate is greater than the preset rate in the preset conditions, the pre-stored digital human system can be started by starting the startup script.

4. The application method of the digital human system according to claim 1, characterized in that, The acquisition of the 3D model data and human motion data of the digital human in the digital human system includes: Obtain the first and second files of the digital human system; In the first file, the three-dimensional model data of the digital human is obtained, and in the second file, the human motion data of the digital human is obtained.

5. The application method of the digital human system according to claim 1, characterized in that, The step of decomposing the 3D model data and the human motion data into multiple processing tasks through a rendering engine, and transmitting the multiple processing tasks to a preset graphics processor, includes: The 3D model data and the human motion data are combined to form rendering data; The rendering engine decomposes the rendering task of the rendering data into multiple processing tasks, and transmits the multiple processing tasks to a preset graphics processor.

6. The application method of the digital human system according to claim 1, characterized in that, After acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device, the method further includes: Collect user voice and convert the user voice into first text information; The first text information is input into a deep learning model to obtain the second text information output by the deep learning model based on the first text information. The second text information is converted into a response voice, and the response voice is sent to the display device.

7. The application method of the digital human system according to claim 6, characterized in that, Inputting the first text information into a deep learning model to obtain second text information output by the deep learning model based on the first text information includes: Obtain the model file of the deep learning model, and load the deep learning model from the model file using a loading script; The first text information is input into the deep learning model to obtain the second text information output by the deep learning model based on the first text information.

8. The application method of the digital human system according to claim 1, characterized in that, After acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device, the method further includes: Obtain an update command, retrieve update content from the server using the update command, perform an update operation on the digital human system using the update content, and obtain the updated digital human system.

9. The application method of the digital human system according to claim 1, characterized in that, After acquiring the digital human image generated by the graphics processor based on multiple processing tasks and sending the digital human image to the display device, the method further includes: Obtain a push command, execute the push command, and push the digital human image to a preset database.

10. A digital human system application device, characterized in that, include: The connection module is used to connect to the display device and start the pre-stored digital human system via a startup script; The acquisition module is used to acquire the 3D model data and human motion data of the digital human in the digital human system; The transmission module is used to decompose the 3D model data and the human motion data into multiple processing tasks through the rendering engine, and transmit the multiple processing tasks to a preset graphics processor. The sending module is used to acquire the digital human image generated by the graphics processor based on multiple processing tasks, and send the digital human image to the display device.

Images