CN115471599A

CN115471599A - Digital human rendering method and system under condition of low-configuration display card

Info

Publication number: CN115471599A
Application number: CN202211206696.4A
Authority: CN
Inventors: 张洽钿; 杜冀中
Original assignee: Beijing Zhipu Huazhang Technology Co ltd
Current assignee: Beijing Zhipu Huazhang Technology Co ltd
Priority date: 2022-09-30
Filing date: 2022-09-30
Publication date: 2022-12-13

Abstract

The application provides a digital human rendering method and a system under the condition of low-configuration display cards, wherein the method comprises the following steps: analyzing the current digital human rendering task, and configuring a corresponding virtual memory for the digital human rendering task according to an analysis result; receiving rendering configuration information and texts of the digital people, generating audio data according to the texts, and generating facial behavior data and limb behavior data of the digital people based on the audio data; issuing an instruction for executing a digital human rendering task to a rendering engine, controlling the rendering engine to execute the digital human rendering task by using a virtual memory, and rendering the digital human according to the facial behavior data and the limb behavior data; and receiving the video stream of the rendered digital person, which is output by the rendering engine, and applying the video stream to a corresponding service scene. The method can output the digital human rendering result with reliable quality under the condition of low configuration of the display card, thereby reducing the rendering cost of the digital human.

Description

Digital human rendering method and system under condition of low-configuration display card

Technical Field

The application relates to the technical field of digital people, in particular to a digital people rendering method and system under the condition of low configuration of a display card.

Background

With the development of Artificial Intelligence (AI) technology, digital people have been widely used in various technical fields. The digital human is a product of the integration of information science and life science, and is used for virtually simulating the shapes and functions of the human body at different levels by using an information science method and realizing the accurate simulation of the human body from microcosmic to macroscopic by establishing a multi-level digital model. By generating a video containing rendered digital people, a variety of functions can be implemented, such as providing information and voice interaction to a user, which requires rendering the digital people prior to applying the digital people.

In the related art, a rendering digital person generally adopts a technical implementation manner of promoting video generation, which includes the following three schemes: firstly, generating a video by pre-generating the video and keying; secondly, generating a scheme based on a preset key image frame; third, an image rendering + compositing based approach to virtual short video generation. However, the applicant finds that, in the above-mentioned solutions in the related art, rendering of a digital person, especially rendering of a super-realistic digital person, is performed by relying heavily on a highly configured graphics card, and rendering cannot be performed without the highly configured graphics card, which results in higher cost and stronger limitation of digital person rendering.

Disclosure of Invention

The present application is directed to solving, at least in part, one of the technical problems in the related art.

Therefore, a first objective of the present application is to provide a digital human rendering method under a low-configuration graphics card condition, where the method configures a corresponding virtual memory for a digital human rendering task, so that a digital human rendering result with reliable quality can be output under the low-configuration graphics card condition, and the cost and the limitation condition of digital human rendering are reduced.

A second objective of the present application is to provide a digital human rendering system under the condition of low-profile graphics card.

A third object of the present application is to propose a non-transitory computer-readable storage medium.

In order to achieve the above object, an embodiment of a first aspect of the present application provides a digital human rendering method under a low-profile graphics card condition, including the following steps:

analyzing a current digital human rendering task, and configuring a corresponding virtual memory for the digital human rendering task according to an analysis result;

receiving rendering configuration information and texts of a digital person, generating audio data according to the texts, and generating facial behavior data and limb behavior data of the digital person based on the audio data;

issuing an instruction for executing the digital human rendering task to a rendering engine, controlling the rendering engine to execute the digital human rendering task by using the virtual memory, and rendering the digital human according to the facial behavior data and the limb behavior data;

and receiving the video stream of the rendered digital person, which is output by the rendering engine, and applying the video stream to a corresponding service scene.

Optionally, in an embodiment of the present application, controlling the rendering engine to execute the digital human rendering task by using the virtual memory includes: judging whether the residual capacity of the virtual memory can meet the digital human rendering task or not in the rendering process; and under the condition that the digital human rendering task is not satisfied, expanding the capacity of the virtual memory.

Optionally, in an embodiment of the present application, after the receiving a video stream of a digital person whose content output by the rendering engine is rendered, the method further includes: detecting the quality of each frame of picture in the video stream, and detecting whether the quality of each frame of picture meets the requirement; and under the condition that the quality of any frame of picture does not meet the requirement, re-rendering is carried out until the quality of each frame of picture meets the requirement.

Optionally, in an embodiment of the present application, the performing quality detection on each frame of picture in the video stream includes: and comparing each frame of picture with a scene reference frame, detecting whether each frame of picture has an abnormality, and determining the position and the number of the abnormality.

Optionally, in an embodiment of the present application, the rendering configuration information includes avatar information and voice style of the digital person, and before the rendering the digital person according to the facial behavior data and the limb behavior data, the method further includes: configuring the image of the digital person according to the image information, wherein the image information comprises the face, the clothes, the hair style and the accessories of the digital person; the rendering digital person further comprises: rendering the avatar of the digital person according to the avatar configuration of the digital person.

Optionally, in an embodiment of the present application, generating audio data according to the text includes: and determining the voice content of the audio data according to the text, and generating the audio data by combining the voice style.

In order to achieve the above object, a second aspect of the present application provides a digital human rendering system under a low-profile graphics card condition, including the following modules:

the configuration module is used for analyzing the current digital human rendering task and configuring a corresponding virtual memory for the digital human rendering task according to an analysis result;

the generating module is used for receiving rendering configuration information and texts of digital people, generating audio data according to the texts, and generating facial behavior data and limb behavior data of the digital people based on the audio data;

the rendering module is used for issuing an instruction for executing the digital human rendering task to a rendering engine, controlling the rendering engine to execute the digital human rendering task by using the virtual memory, and rendering the digital human according to the facial behavior data and the limb behavior data;

and the application module is used for receiving the video stream of the rendered digital person, which is output by the rendering engine, and applying the video stream to a corresponding service scene.

Optionally, in an embodiment of the present application, the rendering module is specifically configured to: judging whether the residual capacity of the virtual memory can meet the digital human rendering task or not in the rendering process; and under the condition that the digital human rendering task is not satisfied, expanding the capacity of the virtual memory.

Optionally, in an embodiment of the present application, the system further includes: a detection module, the detection module specifically configured to: detecting the quality of each frame of picture in the video stream, and detecting whether the quality of each frame of picture meets the requirement; and under the condition that the quality of any frame of picture does not meet the requirement, re-rendering is carried out until the quality of each frame of picture meets the requirement.

In order to implement the foregoing embodiments, an embodiment of the third aspect of the present application further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for digital human rendering under a low-configuration graphics card condition in the foregoing embodiments.

The technical scheme provided by the embodiment of the application at least has the following beneficial effects: according to the method and the device, the corresponding virtual memory is configured for the digital man rendering task, so that the rendering engine performs digital man rendering by using the virtual memory, and the sufficient memory space for rendering resource use can be ensured by using the system virtual memory, so that the rendering of various types of digital people such as 2D, 3D or super-realistic digital people can be completed under the condition of low configuration of the display card. Moreover, the quality of the output rendering result can be ensured by carrying out picture quality detection on the output video stream. Therefore, the digital human rendering result with reliable quality can be output under the condition of low configuration of the display card, and the cost and the limitation of digital human rendering are reduced.

Additional aspects and advantages of the invention will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the invention.

Drawings

The above and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which

Fig. 1 is a flowchart of a digital human rendering method under a low-profile graphics card condition according to an embodiment of the present application;

fig. 2 is a flowchart of a specific digital human rendering method under a low-profile graphics card condition according to an embodiment of the present application;

fig. 3 is a schematic structural diagram of a digital human rendering system under a low-profile graphics card according to an embodiment of the present disclosure.

Detailed Description

Reference will now be made in detail to embodiments of the present invention, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are illustrative and intended to be illustrative of the invention and are not to be construed as limiting the invention.

It should be noted that, in the related art, for a video generation task, a digital person rendering technology greatly depends on a highly configured display card, and is generally limited by a preset video or a key frame image, and cannot provide a complete rich content rendered by a digital person, and the performance of the rendered digital person is not purposefully improved. Therefore, the digital human rendering method under the condition of the low configuration display card can output the digital human rendering result with reliable quality under the condition of the low configuration display card, reduces the digital human rendering cost and reduces the rendering limit.

The following describes a digital human rendering method and system under a low-profile graphics card condition according to an embodiment of the present invention with reference to the accompanying drawings.

Fig. 1 is a flowchart of a digital human rendering method under a low-profile graphics card condition according to an embodiment of the present application, and as shown in fig. 1, the method includes the following steps:

and step S101, analyzing the current digital human rendering task, and configuring a corresponding virtual memory for the digital human rendering task according to an analysis result.

The virtual memory is a technology for managing the memory of the computer system, and the rendering program considers that the rendering program has continuous available memory by configuring the virtual memory.

Specifically, according to the method and the device, the prepositive configuration is carried out before the rendering, the virtual memory is configured for the application program for rendering, and the problem of insufficient memory is solved when the digital human rendering is carried out under the condition that the system is loaded with the low-configuration display card. When the virtual memory is configured, the current digital human rendering task is analyzed, and the corresponding virtual memory is configured for the current digital human rendering task according to the analysis result.

In one embodiment of the application, analyzing the current digital human rendering task includes analyzing the type of the digital human to be rendered for the current rendering task, predicting the rendering data volume of the task, and the like. It will be appreciated that the memory space required for rendering 2D, 3D or super-realistic digital people is increasing, and will vary when different rendering tasks require different amounts of data to be processed. In order to ensure that the configured virtual memory can meet the requirements of rendering different types of digital people, different virtual memory spaces are configured for different digital people rendering tasks, the capacity of the virtual memory configured for the current digital people rendering task is determined by analyzing the types of the digital people to be rendered for the received rendering task, configuration parameters input by a front end and the like, and the capacity of the virtual memory corresponding to the current analysis result can be determined specifically according to historical operating data, expert knowledge and the like.

In specific implementation, the virtual memory with the determined capacity size may be configured by using a physical memory fragment partitioned into a plurality of physical memory fragments in the computer system, an external disk storage, and the like.

And S102, receiving rendering configuration information and texts of the digital person, generating audio data according to the texts, and generating facial behavior data and limb behavior data of the digital person based on the audio data.

In the embodiment of the present application, the text is content of a voice that the rendered digital person needs to make, for example, when the rendered digital person is used for a scene of a customer, the text is content of "hello" or "welcome" that the rendered digital person needs to say. The rendering configuration information is attribute information of the digital person, including image information and voice style of the digital person, and the like.

When the method is specifically implemented, rendering configuration information and texts of digital people input by a front end are received, and audio data are generated according to the texts. As a possible implementation manner, when generating the audio data, the voice content of the audio data is determined according to the content of the text, and then the audio data is generated by combining the voice style. Wherein the voice styles include: in this embodiment, after determining the content to be spoken, a preset AI voice library may be called, the collected sound materials may be obtained, and the sound materials corresponding to the voice style may be screened out to constitute audio data.

Further, facial behavior data and limb behavior data of the digital person are generated based on the audio data. In one embodiment of the present application, facial behavior data and behavior data of limbs may be generated from audio content using an AI model. For example, the facial behavior data includes facial expressions and mouth shapes of the digital person, mouth shapes of the digital person when the digital person speaks words in the content are simulated according to the audio content, and facial expressions of the digital person according to the current scene, for example, when the audio content is "hello", the facial expressions are smile, and the like. The body behavior data includes the body movements of the digital person that fit the current scene, and with continued reference to the above example, when the audio content is "hello", the body behavior data may be hand waving, jumping, and the like.

Furthermore, in one embodiment of the present application, before the rendering is started, the digital human character may be configured according to the rendering configuration information. In this embodiment, the image information of the digital person includes a face, a garment, a hair style, an accessory, and the like of the digital person, and the required image of the digital person is configured according to the received image information of the digital person, so that the digital person image can be conveniently rendered according to the configuration in the subsequent rendering process.

It should be noted that, in the present application, the step of configuring the digital human figure may also be performed before generating the behavior data, and the operation sequence of the two steps may be determined according to the configuration requirement, which is not limited herein.

Therefore, the method and the device determine the face behavior data and the limb behavior data of the digital person to be rendered and the image configuration of the digital person, and improve the vividness and the simulation accuracy of the digital person to be rendered subsequently in a mode of generating the behavior data and the configuration image.

Step S103, sending an instruction for executing the digital human rendering task to the rendering engine, controlling the rendering engine to execute the digital human rendering task by using the virtual memory, and rendering the digital human according to the facial behavior data and the limb behavior data.

The rendering engine may be a 3D modeling rendering model in the related art, such as a UE4 engine and a unity engine.

Specifically, the rendering operation is completed through the rendering engine, the rendering is started after the control instruction is issued to the rendering engine, the facial behavior data and the limb behavior data generated in the above steps are sent to the rendering engine, and then the rendering engine renders the facial expression of the digital human image by using the received facial data and renders the limb actions of the digital human image, including the limbs, the head, the trunk and the like, by using the limb behavior data. The implementation process of rendering by the rendering engine may refer to an implementation manner in the related art, and details are not described here.

When the rendering engine performs rendering, the rendering engine executes a rendering task by using a pre-configured virtual memory. For example, resource data used in a rendering process of a rendering engine and temporary data generated in each operation step are temporarily stored in a memory, and due to the limited memory of the display card, a rendering mode in the related art, the configured memory of the low-configuration display card cannot meet the requirement of storing data in the memory of the display card during rendering, and the system virtual memory can ensure that the memory space used by rendering resources is sufficient.

It can be understood that, since the virtual memory is preconfigured before generating the behavior data according to the received information, in practical applications, since the spaces of the virtual memory required for different rendering tasks are completely different, there may be an error between the capacity of the preconfigured virtual memory and the capacity of the virtual memory required for the current rendering task. In order to ensure that the configured virtual memory can meet different rendering tasks, in an embodiment of the present application, in a process of controlling a rendering engine to execute a digital human rendering task by using the virtual memory, the process includes: judging whether the residual capacity of the virtual memory can meet the digital human rendering task or not in the rendering process; and under the condition that the digital human rendering task is not satisfied, the capacity of the virtual memory is expanded.

Specifically, as the rendering progress increases during the rendering process, the data stored in the virtual memory increases, and the remaining capacity of the virtual memory also decreases. When the residual capacity of the virtual memory is smaller than a preset capacity threshold, the progress of the rendering task determined by the rendering engine and the estimated required memory space can be obtained, and when the residual capacity of the virtual memory is judged to be in the condition of not supporting the current required memory capacity of the digital human rendering task, the capacity of the virtual memory is expanded until the expanded capacity of the virtual memory can support the current required memory capacity of the rendering task.

Furthermore, in order to improve the applicability and accuracy of the preconfigured virtual memory, in an embodiment of the present application, after the rendering engine completes the rendering task each time, the capacity of the virtual memory used by the rendering task this time, the analysis result of the rendering task this time, and the capacity of the virtual memory initially configured actually can also be stored as posterior knowledge, and the policy of configuring the corresponding virtual memory in advance is adjusted, so that the more accurate capacity of the virtual memory can be directly configured when digital human rendering is performed subsequently.

And step S104, receiving the video stream of the rendered digital person, which is output by the rendering engine, and applying the video stream to a corresponding service scene.

Specifically, after the rendering engine completes the rendering task, it may output a video stream, a video file, or a video sequence frame, where the video stream includes a plurality of picture frames, and digital people in the video speak the specified content and make corresponding actions in a corresponding visual simulation. And then receiving the video stream output by the rendering engine, and applying the video stream to a corresponding service scene.

In an embodiment of the present application, in order to ensure the quality of the output rendering result, picture quality detection may be performed on each frame of picture in the video stream output by the rendering engine, and a related image detection algorithm is used to detect whether the picture is normal, that is, whether the quality of each frame of picture meets the requirement, and if the quality of any frame of picture does not meet the requirement, the step S103 is returned to perform rendering again until the quality of each frame of picture meets the requirement.

In this example, whether the quality of the picture meets the requirement includes that there is no abnormality in the picture, for example, the abnormality may be that the image of the digital person in the picture does not meet the configured image, for example, the hair style direction of the digital person is different from the configured image, and the abnormality may also be that there are blank pixel points in the picture that are not successfully rendered.

In specific implementation, the quality detection of each frame of picture in a video stream includes: comparing each frame of picture with a scene reference frame, detecting whether each frame of picture has an abnormality or not, and determining the position and the number of the abnormality. The scene reference frame may be a picture frame having a complete digital human image in a current scene, and other picture frames in the scene are generated based on the reference frame. According to the method and the device, whether the abnormity exists in each frame of picture can be detected in a mode of comparing each frame of picture with a scene reference frame pixel by pixel, the position of the discovered abnormity in the current picture frame is determined through comparison, and the quantity of the abnormity existing in the current picture frame is counted.

Therefore, the quality detection of the output of the rendering result can ensure that the reliable rendering result can be output even if the display card is configured in a low mode.

It should be noted that, in an embodiment of the present application, in order to save computer storage resources, after completing the current task configuration of the digital person rendering, the virtual memory configured for the current task may be deleted, so as to save storage resources for other tasks in the system. And reconfiguring the corresponding virtual memory before executing the digital human rendering task each time.

To sum up, in the digital human rendering method under the condition of low configuration of the graphics card according to the embodiment of the present application, the rendering engine performs digital human rendering by using the virtual memory by configuring the corresponding virtual memory for the digital human rendering task, and the memory space used by rendering resources can be ensured to be sufficient by using the system virtual memory, so that rendering of various types of digital humans of 2D, 3D, or super-realistic digital humans can be completed under the condition of low configuration of the graphics card. And, by performing picture quality detection on the output video stream, the quality of the output rendering result can be ensured. Therefore, the method can output the digital human rendering result with reliable quality under the condition of low configuration of the display card, and reduces the cost and the limitation of digital human rendering.

Based on the above embodiments, in order to more clearly describe a specific processing flow of the digital human rendering method under the condition of the low-configuration graphics card of the present application, in an embodiment of the present application, a specific rendering method is further provided. Fig. 2 is a flowchart of a specific digital human rendering method under a low-profile graphics card condition according to an embodiment of the present disclosure. As shown in fig. 2, the method comprises the steps of:

step S201, configuring a virtual memory for the system.

Step S202, receiving a digital human rendering configuration and a text.

Step S203 generates audio data according to the text content.

Step S204, generating body and facial behavior data based on the audio data.

Step S205, a control instruction is issued to the rendering engine to start rendering.

And step S206, sending the data of the body and face behaviors of the digital person, and rendering the digital person by the rendering engine according to the received data.

Step S207, detecting the picture quality, and determining whether the picture quality is normal, if yes, performing step S208, otherwise, returning to step S205.

In this step, a video stream or a frame output by the rendering engine is received, the picture quality detection is performed, and an image detection algorithm is used to detect whether the picture is normal or not by using the scene reference frame or the adjacent frame as a comparison.

And step S208, inputting the video stream into a service scene for use.

For example, taking a service scene of video output as an example, a video process of "hello" spoken by a digital person is rendered as follows:

firstly, the front end of the system inputs rendering configuration information including digital human images, voice styles and the like; the system starts to generate audio information after receiving the audio information, generates behavior data of the face and the limbs on the basis of the audio information, starts to render and sends the behavior data to drive the digital person after configuring the digital person image, the abnormity detection module detects a picture generated by rendering, controls the rendering engine to render again if abnormity occurs, outputs a video if the abnormal picture is normal, and finally outputs a video that the digital person says 'hello' and makes a corresponding action to return to the front end for displaying.

It should be noted that, for specific implementation of each step in the method, reference may be made to the relevant description of the foregoing embodiment, and details are not described here again.

In order to implement the above embodiments, the present application further provides a digital human rendering system under the condition of low-configuration graphics cards. Fig. 3 is a schematic structural diagram of a digital human rendering system under a low-profile graphics card condition according to an embodiment of the present application.

As shown in fig. 3, the system includes a configuration module 100, a generation module 200, a rendering module 300, and an application module 400.

The configuration module 100 is configured to analyze a current digital human rendering task and configure a corresponding virtual memory for the digital human rendering task according to an analysis result.

The generating module 200 is configured to receive rendering configuration information and text of the digital person, generate audio data according to the text, and generate facial behavior data and limb behavior data of the digital person based on the audio data.

The rendering module 300 is configured to issue an instruction for executing a digital human rendering task to a rendering engine, control the rendering engine to execute the digital human rendering task by using a virtual memory, and render the digital human according to the facial behavior data and the limb behavior data.

The application module 400 is configured to receive a video stream of a digital person whose content is rendered and output by the rendering engine, and apply the video stream to a corresponding service scene.

Optionally, in an embodiment of the present application, the rendering module 300 is specifically configured to: judging whether the residual capacity of the virtual memory can meet the digital human rendering task or not in the rendering process; and under the condition that the digital human rendering task is not satisfied, the capacity of the virtual memory is expanded.

Optionally, in an embodiment of the present application, the system further includes a detection module, specifically configured to: detecting the quality of each frame of picture in the video stream, and detecting whether the quality of each frame of picture meets the requirement; and under the condition that the quality of any frame of picture does not meet the requirement, re-rendering is carried out until the quality of each frame of picture meets the requirement.

Optionally, in an embodiment of the present application, the detection module is specifically configured to: comparing each frame of picture with a scene reference frame, detecting whether each frame of picture has an abnormality or not, and determining the position and the number of the abnormality.

Optionally, in an embodiment of the present application, the rendering configuration information includes avatar information and voice style of the digital person, and the rendering module 300 is further configured to: configuring the image of the digital person according to image information, wherein the image information comprises the face, the clothes, the hair style and accessories of the digital person; and rendering the image of the digital person according to the image configuration of the digital person.

Optionally, in an embodiment of the present application, the generating module 200 is specifically configured to: and determining the voice content of the audio data according to the text, and generating the audio data by combining the voice style.

It should be noted that the foregoing description of the embodiment of the digital human rendering method under the condition of low-profile graphics card is also applicable to the system of the embodiment, and the implementation principle is the same, and is not repeated here.

To sum up, the digital person rendering system under the condition of the low-configuration graphics card in the embodiment of the application configures the corresponding virtual memory for the digital person rendering task, so that the rendering engine performs digital person rendering by using the virtual memory, and can ensure that the memory space used by rendering resources is sufficient by using the system virtual memory, thereby rendering various types of digital persons of 2D, 3D or super-realistic digital persons under the condition of the low-configuration graphics card. Moreover, the quality of the output rendering result can be ensured by carrying out picture quality detection on the output video stream. Therefore, the system can output digital human rendering results with reliable quality under the condition of low configuration of the display card, and reduces the cost and the limitation of digital human rendering.

In order to implement the foregoing embodiments, the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the method for digital human rendering under low-configuration graphics card conditions described in the embodiment of the first aspect of the present application.

In the description of the present specification, reference to the description of "one embodiment," "some embodiments," "an example," "a specific example," or "some examples" or the like means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the present application. In the present specification, if a schematic expression of the above-described terms is employed in a plurality of embodiments or examples, it does not mean that the embodiments or examples are the same. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.

Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless explicitly specified otherwise.

Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present application.

The logic and/or steps represented in the flowcharts or otherwise described herein, such as an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.

It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, various steps or methods may be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.

It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.

In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.

The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. While embodiments of the present application have been shown and described above, it will be understood that the above embodiments are exemplary and should not be construed as limiting the present application and that changes, modifications, substitutions and alterations in the above embodiments may be made by those of ordinary skill in the art within the scope of the present application.

Claims

1. A digital human rendering method under the condition of a low-configuration display card is characterized by comprising the following steps of:

2. The method of digital human rendering according to claim 1, wherein said controlling the rendering engine to perform the digital human rendering task using the virtual memory comprises:

judging whether the residual capacity of the virtual memory can meet the digital human rendering task or not in the rendering process;

and under the condition that the digital human rendering task is not satisfied, expanding the capacity of the virtual memory.

3. The method of claim 1, further comprising, after said receiving the video stream of the rendered digital person as the content output by the rendering engine, the steps of:

detecting the quality of each frame of picture in the video stream, and detecting whether the quality of each frame of picture meets the requirement or not;

and under the condition that the quality of any frame of picture does not meet the requirement, re-rendering is carried out until the quality of each frame of picture meets the requirement.

4. The digital human rendering method of claim 3, wherein the quality detecting each frame of picture in the video stream comprises:

and comparing each frame of picture with a scene reference frame, detecting whether each frame of picture has an abnormality, and determining the position and the number of the abnormality.

5. The digital person rendering method according to claim 1, wherein the rendering configuration information includes character information and voice style of the digital person, and further comprising, before the rendering of the digital person according to the facial behavior data and the body behavior data:

configuring the image of the digital person according to the image information, wherein the image information comprises the face, the clothes, the hair style and the accessories of the digital person;

the rendering digital person further comprises: rendering the avatar of the digital person according to the avatar configuration of the digital person.

6. The digital human rendering method of claim 5, wherein the generating audio data from the text comprises:

and determining the voice content of the audio data according to the text, and generating the audio data by combining the voice style.

7. A digital human rendering system in a low-profile graphics card condition, comprising:

the generating module is used for receiving rendering configuration information and texts of the digital people, generating audio data according to the texts, and generating facial behavior data and limb behavior data of the digital people based on the audio data;

8. The digital human rendering system of claim 7, wherein the rendering module is specifically configured to:

9. The digital human rendering system of claim 7, further comprising: a detection module, the detection module specifically configured to:

10. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements the method for digital human rendering with low-profile graphics card as claimed in any one of claims 1 to 6.