Virtual environment configuration parameter adjusting method and device, electronic equipment and medium

Abstract

The embodiment of the invention provides a virtual environment configuration parameter adjusting method and device, electronic equipment and a medium, and relates to the technical field of three-dimensional scanning modeling. A plurality of virtual light sources, at least one virtual object and a plurality of virtual cameras are arranged in a virtual environment; rendering at least one virtual object through a plurality of virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object; and according to the three-dimensional model data, adjusting configuration parameters of a virtual environment after the plurality of virtual light sources, the plurality of virtual cameras and the at least one virtual object are set. According to the method, the three-dimensional model data of the virtual object is obtained, the configuration of a multi-light-source multi-view-angle shooting system can be evaluated subsequently based on the three-dimensional model data, the configuration can be optimized before the light Stage is actually built, the difficulty and the cost of actually building the light Stage are reduced, the actually built light Stage meets the requirements of a user, and the user experience is improved. And parallel development of hardware and software of the Light Stage is facilitated.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims priority to U.S. application No. 63 / 506,875 filed on June 8, 2023 and entitled “Multiview Capturing Simulation System with Configurable Lightening and Camera,” which application will be incorporated by reference into this application. Technical Field

[0003] The present application relates to the technical field of three-dimensional scanning modeling, and more specifically, to a method, device, electronic device and medium for adjusting configuration parameters of a virtual environment. Background Art

[0004] The multi-light source multi-view shooting system (Light Stage) is a high-fidelity 3D acquisition and reconstruction platform system, which is widely used in the computer vision and computer graphics industry and academia. Light Stage can be used to capture multi-view videos and 3D models of objects. However, the complex structure of Light Stage leads to high cost and difficulty in building Light Stage, and it is impossible to ensure that the final Light Stage meets the actual needs of users. Summary of the invention

[0005] The embodiments of the present application provide a method, device, electronic device and medium for adjusting virtual environment configuration parameters to improve the above-mentioned problems.

[0006] In a first aspect, an embodiment of the present application provides a method for adjusting configuration parameters of a virtual environment, the method comprising: setting a plurality of virtual light sources in a virtual environment, the plurality of virtual light sources being configured to be arranged in a spherical array and facing a target position, the target position being located within the spherical array; setting at least one virtual object in the virtual environment, the at least one virtual object being located at the target position; setting a plurality of virtual cameras in the virtual environment; rendering the at least one virtual object through the plurality of virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object; and adjusting the configuration parameters of the virtual environment after setting the plurality of virtual light sources, the plurality of virtual cameras, and the at least one virtual object according to the three-dimensional model data.

[0007] In a second aspect, an embodiment of the present application provides a device for adjusting configuration parameters of a virtual environment, the device comprising: a light source simulation module, used to set a plurality of virtual light sources in a virtual environment, the plurality of virtual light sources being configured to be arranged in a spherical array and facing a target position, the target position being located within the spherical array; an object simulation module, used to set at least one virtual object in the virtual environment, the at least one virtual object being located at the target position; a camera simulation module, used to set a plurality of virtual cameras in the virtual environment; an object rendering module, used to render the at least one virtual object through the plurality of virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object; and a parameter adjustment module, used to adjust the configuration parameters of the virtual environment after setting the plurality of virtual light sources, the plurality of virtual cameras and the at least one virtual object according to the three-dimensional model data.

[0008] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a memory and a processor, wherein an application is stored in the memory, and the processor executes the method provided by the embodiment of the present application when calling the application.

[0009] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having program code stored thereon, and a processor executes the method provided in the embodiment of the present application when calling the program code.

[0010] The method for adjusting the configuration parameters of the virtual environment provided in the embodiment of the present application can virtually simulate a multi-light source and multi-angle shooting system (Light Stage) by setting multiple virtual light sources, at least one virtual object, and multiple cameras in the virtual environment. By rendering at least one virtual object through multiple virtual cameras, the three-dimensional model data corresponding to the at least one virtual object can be obtained, so that the configuration of the Light Stage can be evaluated based on the three-dimensional model data in the future, which helps to optimize the configuration of the Light Stage before actually building the Light Stage, reduce the difficulty and cost of actually building the Light Stage, and make the actually built Light Stage meet user needs. In addition, by generating three-dimensional model data based on the simulated virtual Light Stage, the performance of the software module of the Light Stage can be verified, so as to facilitate the parallel development of the hardware and software of the Light Stage. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following is a brief introduction to the drawings required for the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments and drawings obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0012] Figure 1 shows a grayscale image of a real Light Stage in the related art;

[0013] Figure 2 A schematic diagram showing a flow chart of a method for adjusting virtual environment configuration parameters provided by an embodiment of the present application;

[0014] Figure 3 shows a grayscale image of multiple virtual light sources provided by an exemplary embodiment of the present application;

[0015] Figure 4 A schematic diagram of a configuration interface provided by an exemplary embodiment of the present application is shown;

[0016] Figure 5 A virtual camera pair provided by an exemplary embodiment of the present application is shown. Figure 4 A grayscale image of the image rendered by the human body model in ;

[0017] Figure 6 Another virtual camera pair provided by an exemplary embodiment of the present application is shown. Figure 4 A grayscale image of the image rendered by the human body model in ;

[0018] Figure 7 Another virtual camera pair provided by an exemplary embodiment of the present application is shown. Figure 4 A grayscale image of the image rendered by the human body model in ;

[0019] Figure 8 shows a grayscale image of multiple virtual cameras provided by an exemplary embodiment of the present application;

[0020] Fig. 9 A virtual camera pair provided by another exemplary embodiment of the present application is shown. Figure 8 A grayscale image of the image rendered by the human body model in ;

[0021] Fig.10 Another virtual camera pair provided by another exemplary embodiment of the present application is shown. Figure 8 A grayscale image of the image rendered by the human body model in ;

[0022] Fig.11A grayscale image of an image obtained by rendering a virtual object under a first lighting condition by a virtual camera provided by another exemplary embodiment of the present application is shown;

[0023] Fig.12 A grayscale image of an image obtained by rendering a virtual object under a second lighting condition by a virtual camera provided by another exemplary embodiment of the present application is shown;

[0024] Fig.13 A schematic diagram showing a flow chart of a method for adjusting virtual environment configuration parameters provided by another embodiment of the present application is shown;

[0025] Fig.14 shows a grayscale image of one of the images in the reverse gradient illumination image pair provided by an exemplary embodiment of the present application;

[0026] Fig.15 shows a grayscale image of another image in a pair of reverse gradient illumination images provided by an exemplary embodiment of the present application;

[0027] Fig.16 An exemplary embodiment of the present application provides Fig.14 and Fig.15 A grayscale image of the normal estimation image obtained as input;

[0028] Fig.17 An exemplary embodiment of the present application provides Fig.14 and Fig.15 Grayscale image of the normal truth image obtained as input;

[0029] Fig.18 Another exemplary embodiment of the present application provides Fig.11 and Fig.12 A grayscale image of the normal estimation image obtained as input;

[0030] Fig.19 Another exemplary embodiment of the present application provides Fig.11 and Fig.12 Grayscale image of the normal truth image obtained as input;

[0031] Fig. 20 A schematic diagram showing a normal estimation error curve of an image rendered under different lighting environments provided by an exemplary embodiment of the present application is shown;

[0032] Fig.21 A grayscale image of an image rendered based on albedo provided by an exemplary embodiment of the present application is shown;

[0033] Fig. 22 A grayscale image of an image rendered based on specular reflectivity provided by an exemplary embodiment of the present application is shown;

[0034] Fig.23 A grayscale image of an image rendered based on normals provided by an exemplary embodiment of the present application is shown;

[0035] Fig.24 It shows that an embodiment of the present application provides Figures 5 to 7 Grayscale image of the normal truth image obtained by photometric stereo imaging;

[0036] Fig.25 A grayscale image of a virtual object partially surrounded by four virtual cameras provided by an exemplary embodiment of the present application is shown;

[0037] Fig.26 A grayscale image showing six virtual cameras completely surrounding a virtual object provided by an exemplary embodiment of the present application is shown;

[0038] Fig. 27 A schematic diagram showing a flow chart of a method for adjusting virtual environment configuration parameters provided by another embodiment of the present application;

[0039] Fig.28 A schematic diagram of the structure of a device for adjusting virtual environment configuration parameters provided in an embodiment of the present application is shown;

[0040] Fig.29 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0041] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0042] See also Figure 1 , Figure 1 FIG. 1 shows a grayscale image of a real Light Stage in the related art. Figure 1 As shown, LightStage uses a camera array composed of multiple programmable light sources and multiple high-precision cameras. The objects to be photographed are set inside the Light Stage, which can realize multi-light source and multi-angle shooting of objects in the Light Stage, restore the three-dimensional structure of the photographed objects, and obtain the reflection information on the objects, so as to obtain the light effect of reconstructing the three-dimensional model of the object under different ambient lights. Light Stage can control the lighting direction, lighting intensity and spectral color of the light source, and has a wide range of applications in many fields. For example, in the field of film and video, Light Stage can be used to obtain high-resolution human faces and obtain very high-resolution face models in a large number of different facial expressions.

[0043] However, the inventors found in their research that the Light Stage is complex in both hardware and software, and needs to be carefully designed before starting to build the Light Stage. On the one hand, the Light Stage has a complex structure and is difficult to build. It is difficult to build a Light Stage that satisfies the user in one go. In the process of building the Light Stage, potential back-and-forth efforts and expenses will be involved due to construction errors or incorrect hardware or insufficient configuration, resulting in increased construction costs. On the other hand, before the Light Stage is built, the algorithms or software modules involved in the Light Stage cannot be verified, and the hardware and software must be developed in sequence, that is, the Light Stage must be built first and then the algorithms or software modules of the Light Stage must be developed, rather than developed in parallel, resulting in lower development efficiency of the Light Stage.

[0044] Based on the above analysis, the inventor proposed a method for adjusting the configuration parameters of a virtual environment. By setting multiple virtual light sources, at least one virtual object and multiple cameras in a virtual environment, a light stage is virtually simulated. Before building a real light stage, the configuration of the light stage can be evaluated and the algorithm of the light stage can be verified based on the virtually simulated light stage, thereby saving the cost and difficulty of building a real light stage and improving the development efficiency of the light stage.

[0045] The method for adjusting the virtual environment configuration parameters in the embodiment of the present application can be applied to electronic devices, in which a three-dimensional creation engine (software) is deployed. The three-dimensional creation engine may include but is not limited to Blender, Unity, Unreal Engine (UE for short), 3D MAX, and Maya, etc.

[0046] See also Figure 2 , Figure 2 A flow chart of a method for adjusting virtual environment configuration parameters provided by an embodiment of the present application is shown. The method for adjusting virtual environment configuration parameters may include steps S110 to S150.

[0047] Step S110: multiple virtual light sources are set in the virtual environment, and the multiple virtual light sources are configured to be arranged in a spherical array and face a target position, and the target position is located in the spherical array.

[0048] The virtual environment is relative to the real environment. The virtual environment can be created in a three-dimensional creation engine first, so that relevant content (such as virtual light sources, virtual objects, and virtual cameras) can be added to the virtual environment later. It can be understood that the virtual environment is three-dimensional.

[0049] The virtual light source in the embodiment of the present application can be changed dynamically. For example, its lighting intensity, lighting color, lighting direction, etc. can all be changed dynamically, thereby simulating different lighting environments and lighting conditions, so as to facilitate the subsequent capture of three-dimensional model data of virtual objects in different lighting environments.

[0050] The target position in the embodiment of the present application refers to the position inside the spherical array for placing the virtual object, for example, the target position may be the center of the spherical array. It is understandable that the target position and the size of the spherical array may be set according to actual needs.

[0051] In some embodiments, a virtual spherical array can be created in a virtual environment, and multiple virtual light sources can be set on the spherical array. By creating the spherical array first and then setting the virtual light sources, the difficulty of setting the virtual light sources can be reduced and the efficiency of setting the virtual light sources can be improved.

[0052] In other embodiments, multiple virtual light sources arranged in a spherical array may be directly set in the virtual environment without creating a virtual spherical array. By directly setting the virtual light sources without creating a virtual spherical array, the content in the virtual environment may be reduced and the efficiency of subsequent rendering may be improved.

[0053] In some embodiments, multiple virtual light sources may be evenly arranged in a spherical array, so that the multiple virtual light sources evenly illuminate the virtual object at the target position. Figure 3 , Figure 3 FIG. 2 shows a grayscale image of multiple virtual light sources provided by an exemplary embodiment of the present application. Figure 3 As shown, the white dots represent virtual light sources, and the position of ① is the target position, for example, it can be the center position of a spherical array composed of virtual light sources.

[0054] In some embodiments, multiple virtual light sources may be arranged unevenly in a spherical array. For example, the number of virtual light sources at the bottom of the spherical array may be set to be less than the number of virtual light sources at the top and middle of the spherical array, and the density of virtual light sources at the bottom of the spherical array may be set to be sparser than the density of virtual light sources at the top and middle of the spherical array, so as to illuminate key areas of virtual objects at target locations. It is understandable that the position of the virtual light source may be set according to the shape of the virtual object to be rendered.

[0055] In some embodiments, the three-dimensional creation engine has a configuration interface, and the user can set multiple virtual light sources in the virtual environment based on the configuration interface. Setting the configuration interface can reduce the difficulty of the user setting the virtual light source and simplify the user operation.

[0056] In other embodiments, the 3D creation engine has an application programming interface (API). Based on the API, the 3D creation engine can provide users with a code or configuration file writing window, so that users can set multiple virtual light sources in a virtual environment by writing code (such as Python scripts) or writing configuration files (such as Jason configuration files).

[0057] Step S120: setting at least one virtual object in the virtual environment, wherein the at least one virtual object is located at a target position.

[0058] The virtual object in the embodiment of the present application is a three-dimensional object. The virtual object can be custom created by the user based on a three-dimensional creation engine, or can be downloaded from a cloud or shared 3D model database or platform (such as the sketchfab platform).

[0059] In some embodiments, the 3D creation engine has a configuration interface (also referred to as a virtual object editing interface). The target position can be prompted to the user, and the user can set at least one virtual object in the virtual environment based on the configuration interface, and set the position of at least one virtual object based on the target position. For example, see Figure 4 , Figure 4 A schematic diagram of a configuration interface provided by an exemplary embodiment of the present application is shown. Figure 4 As shown, a human model (virtual object) can be set in a virtual environment based on the configuration interface, such as Figure 3 As shown, the human body model can be set at the target position ①.

[0060] In other embodiments, the 3D creation engine has an API, based on which the 3D creation engine can provide the user with a code or configuration file writing window, so that the user can set at least one virtual object in the virtual environment by writing code or writing a configuration file.

[0061] Step S130: Setting a plurality of virtual cameras in the virtual environment.

[0062] In some embodiments, multiple virtual cameras may be arranged in a virtual environment to surround at least one virtual object, and the multiple virtual cameras may all face at least one virtual object (ie, a target location) to facilitate subsequent 360-degree capture and modeling of the at least one virtual object.

[0063] In some embodiments, multiple virtual cameras may be arranged in a virtual environment to be located on one side of at least one virtual object, so as to subsequently capture and model an important side of the at least one virtual object instead of 360 degrees.

[0064] In some embodiments, a configuration interface is provided in the three-dimensional creation engine. A mapping relationship between the size of the spherical array and the position information of multiple virtual cameras can be preset; based on the mapping relationship, multiple virtual camera position information corresponding to the size of the spherical array is automatically generated as the recommended position information of multiple virtual cameras; the target position and the recommended position information of multiple virtual cameras are prompted to the user, and the user can set multiple virtual cameras in the virtual environment based on the configuration interface, and set the orientation and position (i.e., posture) of the multiple virtual cameras based on the target position and the recommended position information of multiple virtual cameras.

[0065] In other embodiments, the 3D creation engine has an API, based on which the 3D creation engine can provide the user with a code or configuration file writing window, so that the user can set up multiple virtual cameras in a virtual environment by writing code or configuration files.

[0066] It is understandable that the multiple virtual cameras may be located outside the spherical array, so as to avoid affecting the lighting of the at least one virtual object by the multiple virtual light sources.

[0067] It can be understood that the execution order of steps S110 to S130 can be changed according to actual needs, that is, the setting order of virtual light sources, virtual objects, and virtual cameras can be flexibly adjusted according to actual needs and does not have a fixed execution order. This application does not limit the execution order of steps S110 to S130.

[0068] Step S140: Render at least one virtual object through multiple virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object.

[0069] The rendering conditions of multiple virtual cameras can be set, and the rendering conditions may include but are not limited to: the resolution of the rendered image, the output storage path, and the imaging quality. All images obtained by multiple virtual cameras are rendered according to the rendering conditions, and this process is the process of simulating camera imaging. The software algorithm of the Light stage can be embedded in the 3D creation engine, and based on all images of multiple virtual cameras, the software algorithm is called to generate 3D model data corresponding to at least one virtual object, thereby realizing 3D capture and 3D model reconstruction of at least one virtual object.

[0070] As an example, see Figures 5 to 7 , Figure 5 A virtual camera pair provided by an exemplary embodiment of the present application is shown. Figure 4 The grayscale image of the image rendered by the human body model in Figure 6 Another virtual camera pair provided by an exemplary embodiment of the present application is shown. Figure 4 The grayscale image of the image rendered by the human body model in Figure 7 Another virtual camera pair provided by an exemplary embodiment of the present application is shown. Figure 4 The grayscale image of the image is rendered by the human body model in . Figures 5 to 7 As shown, multiple virtual cameras surround at least one virtual object (such as Figure 4 When a human body model in is set, multiple virtual cameras can capture images of at least one virtual object at different perspectives.

[0071] As an example, see Figures 8 to 10 , Figure 8 shows a grayscale image of multiple virtual cameras provided by an exemplary embodiment of the present application, Fig. 9 A virtual camera pair provided by another exemplary embodiment of the present application is shown. Figure 8 The grayscale image of the image rendered by the human body model in Fig.10 Another virtual camera pair provided by another exemplary embodiment of the present application is shown. Figure 8 The grayscale image of the image is rendered by the human body model in the figure. Multiple virtual cameras are arranged as follows: Figure 8 When the positions and orientations shown are set around the human body model, Fig. 9 and Fig.10 As shown, multiple virtual cameras can render Figure 8 Images of the human body model at different viewing angles.

[0072] For examples, see Fig.11 and Fig.12 , Fig.11 FIG. 4 shows a grayscale image of an image obtained by rendering a virtual object (3D face model) under a first lighting condition by a virtual camera provided by another exemplary embodiment of the present application. Fig.12 FIG. 1 shows a grayscale image of an image obtained by rendering a virtual object by a virtual camera under a second lighting condition provided by another exemplary embodiment of the present application, wherein the 3D face model is located at Figure 3 The target position in ①. Fig.11 and Fig.12 As shown, Fig.11 The image shown has a light intensity higher than Fig.12The illumination intensity of the image shown shows that different illumination conditions can be simulated by multiple virtual light sources, so that the virtual camera can capture images under different illumination conditions.

[0073] like Figure 5-12 As illustrated by the rendered image shown, the present application can operate like a real light stage based on multiple virtual light sources, at least one virtual object and multiple virtual cameras, and can simulate the three-dimensional capture and modeling of the real light stage, thereby helping relevant personnel to design the light stage before actually building the light stage. The virtual light stage system constructed by the present application can provide flexibility for the design and production of the real light stage.

[0074] Step S150: adjusting configuration parameters of the virtual environment after setting a plurality of virtual light sources, a plurality of virtual cameras and at least one virtual object according to the three-dimensional model data.

[0075] The configuration parameters of the virtual environment may include one or more combinations of configuration parameters of multiple virtual light sources, multiple virtual cameras, and at least one virtual object. For a detailed description of adjusting the configuration parameters of the virtual environment, please refer to steps S250 to S260.

[0076] Based on steps S110 to S150, the present application can virtually simulate a multi-light source and multi-angle shooting system (Light Stage) by setting multiple virtual light sources, at least one virtual object, and multiple cameras in a virtual environment. By rendering at least one virtual object through multiple virtual cameras, three-dimensional model data corresponding to at least one virtual object can be obtained, so that the configuration of the Light Stage can be evaluated based on the three-dimensional model data in the future, which helps to optimize the configuration of the Light Stage before actually building the Light Stage, reduce the difficulty and cost of actually building the Light Stage, and make the actually built Light Stage meet user needs. In addition, by generating three-dimensional model data based on the simulated virtual Light Stage, the performance of the software module of the Light Stage can be verified, so as to facilitate the parallel development of the hardware and software of the Light Stage.

[0077] See also Fig.13 , Fig.13 A flow chart of a method for adjusting virtual environment configuration parameters provided by another embodiment of the present application is shown. The method for adjusting virtual environment configuration parameters may include steps S210 to S260.

[0078] Step S210: multiple virtual light sources are set in the virtual environment, and the multiple virtual light sources are configured to be arranged in a spherical array and face a target position, and the target position is located in the spherical array.

[0079] Step S220: setting at least one virtual object in the virtual environment, wherein the at least one virtual object is located at a target position.

[0080] Step S230: Setting a plurality of virtual cameras in the virtual environment.

[0081] Step S240: Render at least one virtual object through multiple virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object.

[0082] Step S250: adjusting configuration parameters of multiple virtual light sources according to the three-dimensional model data corresponding to at least one virtual object.

[0083] According to the three-dimensional model data corresponding to at least one virtual object (including images rendered by multiple virtual cameras and three-dimensional models established based on the images rendered by multiple virtual cameras), a photometric stereo (PMS) imaging method and / or a de-rendering method can be used to evaluate the configuration parameters of multiple virtual light sources; and according to the evaluation results, the configuration parameters of the multiple virtual light sources are adjusted.

[0084] Photometric stereo is a technique for reconstructing the surface of an object by using multiple images. Edges and textures are generated from the 2D texture or surface shading of the object, that is, images of the object illuminated from multiple directions are taken first, and then a photometric stereo algorithm is used to generate a complete and clear image. Fig.14 and Fig.15 , Fig.14 shows a grayscale image of one of the images in the reverse gradient illumination image pair provided by an exemplary embodiment of the present application, Fig.15 FIG. 2 shows a grayscale image of another image in a pair of reverse gradient illumination images provided by an exemplary embodiment of the present application. Fig.14 and Fig.15 As shown, given a pair of inverse gradient lighting conditions, estimating the three-dimensional surface normal of a virtual object is a key step in photometric stereo imaging. If the virtual object is set at the center of a spherical array, an ideal photometric stereo imaging effect can be obtained. However, in actual operation, it is difficult to meet the above conditions when capturing moving objects, especially large moving objects on the ground. This application is based on a photometric stereo imaging method to quantitatively show the impact if the object is not at the center of a spherical array.

[0085] For example, to render spherical gradient lighting, the diameter of the spherical array can be set to 8 units (e.g., decimeters), and the spherical array can be placed at the target position (0,0,0); multiple virtual light sources can be set to 96 RGB (Red Green Blue) multi-spectral lights, and the 96 RGB multi-spectral lights can be evenly arranged in the spherical array and set to point to the target position (0,0,0), so that for each object position, all light intensity directions are from the light source toward the center of the object; a spherical object (virtual object) with a diameter of 2 units can be set at the target position (0,0,0); multiple virtual cameras can be set around the spherical object. Multiple virtual cameras can render images of the spherical object under different lighting conditions to simulate the capture of the spherical object by a real camera under different lighting conditions.

[0086] In order to evaluate the performance of normal estimation under different spherical gradient illumination settings, the spherical object is moved from the first position of the z-axis of the spherical array (e.g., the position of z=-3) to the second position (e.g., the position of z=3), and all light intensities corresponding to the illumination direction are set at the center of the spherical array as general gradient light. Images are rendered through multiple virtual cameras under spherical gradient illumination, and the surface normals of the rendered images are estimated to obtain normal estimation images.

[0087] As an example, the comparison between the normal estimation image and the normal truth image can be as follows: Fig.16 and Fig.17 As shown, Fig.16 An exemplary embodiment of the present application provides Fig.14 and Fig.15 (Inverse gradient illumination image pair) A grayscale image of the normal estimation image obtained as input; Fig.17 An exemplary embodiment of the present application provides Fig.14 and Fig.15 (Inverse Gradient Illumination Image Pair) Grayscale image of the normal truth image obtained as input.

[0088] As another example, the comparison between the normal estimation image and the normal truth image can be as follows: Fig.18 and Fig.19 As shown, Fig.18 Another exemplary embodiment of the present application provides Fig.11 and Fig.12 (Inverse gradient illumination image pair) is the grayscale image of the normal estimation image obtained as input, Fig.19 Another exemplary embodiment of the present application provides Fig.11 and Fig.12 (Inverse Gradient Illumination Image Pair) Grayscale image of the normal truth image obtained as input.

[0089] For examples, see Fig. 20 , Fig. 20 A schematic diagram of a normal estimation error curve of an image rendered under different lighting environments provided by an exemplary embodiment of the present application is shown, wherein: Fig. 20 The horizontal axis represents the position of the virtual object on the z-axis of the spherical array, and the vertical axis represents the average normal error of the rendered image, which is calculated in degrees. Fig. 20 In FIG. 1 , a relatively thin curve represents an average normal estimation error curve in the process of a virtual object moving from z=-3 of the spherical array to z=3 in the first illumination environment, and a relatively thick curve represents an average normal estimation error curve in the process of a virtual object moving from z=-3 of the spherical array to z=3 in the second illumination environment. Fig. 20 As shown, it can be seen that the configuration parameters (lighting configuration) of the multiple virtual light sources under the second lighting condition are better than those under the first lighting condition, because the configuration parameters of the multiple virtual light sources under the second lighting condition better match the virtual objects (e.g. Fig.14 and Fig.15 The spherical object shown or Fig.11 and Fig.12 Gradient lighting when the face model (shown) moves.

[0090] That is, using the photometric stereo imaging method to evaluate the configuration parameters of multiple virtual light sources may include: obtaining rendering images of multiple virtual cameras during the movement of at least one virtual object (spherical object) from a first z-axis position (e.g., position z=-3) to a second position (e.g., position z=3) of the spherical array; performing normal estimation on the acquired rendering image to obtain a normal estimation error; based on the normal estimation error, evaluating the configuration parameters (e.g., lighting configuration) of multiple virtual light sources; and adjusting the configuration parameters of the multiple virtual light sources according to the evaluation results.

[0091] The de-rendering method is a technique for reverse modeling the image formation process of the captured object and the geometric shape and material properties of the object (such as surface shape, shadow, albedo, normal, reflection, etc.) under given specific lighting conditions. The inference of this property is also called intrinsic image estimation. Among them, the intrinsic image refers to an image composed of multiple images, that is, an image is decomposed into multiple parts, and the multiple images obtained by decomposition are the intrinsic images of the original image. Among them, decomposition into several parts can have different definitions. For example, an image is decomposed into its corresponding reflection map and illumination map (or called true color map and highlight map), or an image can be decomposed into multiple images based on albedo, specular reflectivity, normal (normal vector), surface roughness, subsurface reflectivity, surface shape, shadow, reflectivity and other factors. In other words, the purpose of de-rendering is to decompose the components from the original image that can combine how the original image is formed, and this application does not limit the specific decomposition method.

[0092] For examples, see Figures 5 to 7 as well as Figure 21 to Figure 23 , Fig.21 shows a grayscale image of an image rendered based on albedo provided by an exemplary embodiment of the present application, Fig. 22 FIG. 4 shows a grayscale image of an image rendered based on the specular reflectivity provided by an exemplary embodiment of the present application. Fig.23 FIG. 1 shows a grayscale image of an image rendered based on normals provided by an exemplary embodiment of the present application. Figures 5 to 7 The multiple multi-view images shown are applied to the rendering method, such as Figure 21 to Figure 23 As shown, this application can be restored Figures 5 to 7 Surface texture material properties for multiple multi-view images are shown. Also, see Fig.24 , Fig.24 It shows that an embodiment of the present application provides Figures 5 to 7 Grayscale image of the normal truth image obtained by photometric stereo imaging, Fig.24 The estimated Peak Signal to Noise Ratio (PSNR) of the normal is 33.23, and the Mean Square Error (MSE) is 0.0006.

[0093] That is to say, evaluating the configuration parameters of multiple virtual light sources using a de-rendering method may include: obtaining images rendered by multiple virtual cameras under different lighting conditions (multi-view images), processing the multi-view images using a de-rendering method, restoring the intrinsic properties of virtual objects in the images, and evaluating the error between the restored intrinsic properties and the true value to evaluate the configuration parameters of multiple virtual light sources in the virtual Light Stage (lighting change configuration).

[0094] Combining the photometric stereo imaging method and the de-rendering method, by adjusting different lighting conditions and measuring the normal estimation error under different lighting conditions, the configuration parameters of multiple virtual light sources in the virtual Light Stage can be evaluated to improve the robustness of the virtual Light Stage to lighting changes.

[0095] Step S260: adjusting configuration parameters of multiple virtual cameras according to the three-dimensional model data corresponding to at least one virtual object.

[0096] According to the three-dimensional model data corresponding to at least one virtual object, the configuration parameters of multiple virtual cameras are evaluated by using the Neural Radiance Fields (NeRF) method; according to the evaluation results, the configuration parameters of the multiple virtual cameras are adjusted.

[0097] Among them, the neural radiance field method is used to synthesize images under different perspectives of the same lighting environment. That is, based on several images of a given lighting environment, the 3D representation of objects in the lighting environment is reconstructed, and then a certain perspective is input during reasoning to synthesize (render) the image (including objects) under that perspective.

[0098] In the training and rendering (inference) stages, neural networks can be used to capture the radiation field properties of multi-view images, thereby achieving scene rendering of multi-view images. Instant-NGP can be used as a neural radiation field model for evaluation. Among them, Instant-ngp is mainly used to solve the efficiency problem of neural radiation fields when parameterizing fully connected neural networks. Instant-ngp uses a smaller-scale network to implement neural radiation fields without loss of accuracy, thereby reducing the time overhead of neural radiation fields from hours to seconds.

[0099] This application mainly implements the following two evaluations based on the neural radiance field method: 1. Evaluate the impact of the number of multiple virtual cameras (i.e., the number of views of the input image); 2. Evaluate the minimum number of multiple virtual cameras (i.e., the number of views) for training the neural radiance field model and its performance.

[0100] Table 1

[0101]

[0102]

[0103] For the first case, please refer to Figure 8 , Figure 8 The positions and orientations (poses) of multiple virtual cameras are shown, and the performance of the neural radiation field model can be evaluated from using a first number (e.g., 270) of virtual cameras as input to using a second number (e.g., 27) of virtual cameras as input. Table 1 shows how the number of virtual cameras affects the rendering accuracy based on the peak signal-to-noise ratio metric. As shown in Table 1, the performance of the neural radiation field model tends to degrade as the number of virtual cameras decreases. Evaluating the number of multiple virtual cameras based on the neural radiation field method can provide clear guidance on how to set the number of multiple virtual cameras. Whether it is the training or inference process, when computing resources are limited, adjusting the number of multiple virtual cameras based on the evaluation results of the neural radiation field method helps to strike a balance between quality and speed.

[0104] Table 2

[0105] Iterations Training duration Training PSNR Testing PSNR Training SSIM Testing SSIM 500 5.9218 35.9160 34.9466 0.9466 0.9308 1000 10.7285 35.8959 35.1314 0.9569 0.9414 1500 15.4940 35.8798 35.1596 0.9584 0.9432 2000 20.2117 35.8756 35.1814 0.9580 0.9433 2500 24.8711 35.8504 35.1677 0.9599 0.9449

[0106] Table 3

[0107] Iterations Training duration Training PSNR Testing PSNR Training SSIM Testing SSIM 500 5.7599 35.7366 35.5311 0.9477 0.9461 1000 10.7448 35.7453 35.6539 0.9537 0.9517 1500 15.5346 35.7185 35.6669 0.9560 0.9537 2000 20.2868 35.6954 35.6708 0.9543 0.9520 2500 25.1171 35.6931 35.6812 0.9551 0.9525

[0108] For the second case, see Figure 25 to Figure 26 , Fig.25 FIG. 4 shows a grayscale image of a virtual object (virtual human body model) partially surrounded by four virtual cameras provided by an exemplary embodiment of the present application. Fig.26 A grayscale image showing six virtual cameras completely surrounding a virtual object (virtual human body model) provided by an exemplary embodiment of the present application is shown. Figure 25 to Figure 26 The white solid circles in the figure represent virtual cameras. Evaluating the quality of rendered images and training time given 6 and 4 virtual cameras helps to make the trade-off between cost, latency, and quality. Fig.25 The latency test and quantitative evaluation results of rendering with 4 virtual cameras are shown in Table 2. Fig.26 The latency test and quantitative evaluation results of rendering with 6 virtual cameras are shown in Table 3. The SSIM in Tables 2 and 3 is structural similarity, which is used to measure the similarity between different images. The training time in Tables 2 and 3 is in seconds.

[0109] Combining the evaluation results of the first and second cases, adjusting the configuration parameters of multiple virtual cameras (such as position and quantity) can help flexibly adjust the camera configuration of the Light Stage, which can provide a reference for building a real Light Stage.

[0110] It can be understood that the present application does not limit the specific execution of step S250 and step S260. According to actual needs, one of step S250 and step S260 can be selected for execution, or step S250 and step S260 can be executed serially, or step S250 and step S260 can be executed in parallel.

[0111] Based on step S210 to step S260, the present application can virtually simulate a multi-light source multi-view shooting system (Light Stage) by setting multiple virtual light sources, at least one virtual object and multiple cameras in a virtual environment, and render at least one virtual object through multiple virtual cameras, and obtain the three-dimensional model data corresponding to at least one virtual object, so as to facilitate the subsequent evaluation of the configuration of the Light Stage based on the three-dimensional model data, which helps to optimize the Light Stage configuration before the actual construction of the Light Stage, reduce the difficulty and cost of the actual construction of the Light Stage, and make the actual construction of the Light Stage meet the needs of users. In addition, based on the generation of three-dimensional model data based on the simulated virtual Light Stage, the performance of the software module of the Light Stage can be verified, so as to facilitate the parallel development of the hardware and software of the Light Stage. In addition, the configuration parameters (lighting configuration) of multiple virtual light sources are evaluated in combination with the photometric stereo imaging method and the de-rendering method, and the configuration parameters (camera configuration) of multiple virtual cameras are evaluated based on the neural radiation field method. The lighting configuration and / or camera configuration are adjusted based on the evaluation results, so as to improve the flexibility of the Light Stage design, and the optimal configuration of the Light Stage can be achieved based on multiple evaluations and adjustments, so as to provide a reference for building a real Light Stage.

[0112] See also Fig. 27 , Fig. 27 A flowchart of a method for adjusting virtual environment configuration parameters provided by another embodiment of the present application is shown. The method for adjusting virtual environment configuration parameters may include steps S310 to S350.

[0113] Step S310: Detecting configuration operations performed by a user on a configuration interface, where the configuration interface is used to configure configuration parameters of multiple virtual light sources, at least one virtual object, and multiple virtual cameras.

[0114] The configuration parameters of the multiple virtual light sources, at least one virtual object, and the multiple virtual cameras may include at least one or a combination of parameters such as quantity, position, shape, direction, size, and operation mode parameters. In some embodiments, the configuration parameters of the multiple virtual light sources include operation mode parameters, and the operation mode parameters include a dynamic change mode and a lighting intensity and color corresponding to the dynamic change mode. Different dynamic change modes correspond to different lighting intensities and colors.

[0115] The 3D creation engine interface includes at least a display interface and a configuration interface. The display interface is used to display the virtual environment and its content. The content in the virtual environment may include multiple virtual light sources, at least one virtual object, and multiple virtual cameras. The configuration interface includes multiple sub-interfaces, and different sub-interfaces are used to configure different configuration parameters. For example, the sub-interface related to the virtual light source is used to configure the configuration parameters corresponding to the virtual light source, the sub-interface related to the virtual object is used to configure the configuration parameters corresponding to the virtual object, and the sub-interface related to the virtual camera is used to configure the configuration parameters corresponding to the virtual light source.

[0116] In some embodiments, each sub-interface of the configuration interface provides multiple selection items (such as a slide bar or a drop-down bar) and / or input items (such as a click input item), and the user can perform configuration operations in the configuration interface based on the multiple selection items and / or input items, that is, the configuration operation may include but is not limited to a click operation, a selection operation, an input operation, etc. In response to detecting a configuration operation, the three-dimensional creation engine can obtain virtual environment configuration information corresponding to the configuration operation, and perform operations corresponding to the configuration operation in the virtual environment based on the virtual environment configuration information, so that the user can see the configuration results corresponding to his or her configuration operation in real time, which can improve configuration efficiency and user experience during configuration. For example, the user can input the number of virtual light sources as 96 in the configuration interface, and the three-dimensional creation engine can obtain virtual environment configuration information that the number of virtual light sources is 96 in response to the input operation, and set 96 virtual light sources in the virtual environment.

[0117] In other embodiments, the configuration interface may also include an API, and the API may provide a code or configuration file writing window for the user. The user may open the API in the configuration interface, and the three-dimensional creation engine may display a code or configuration file writing window corresponding to the API. The user may write code or a configuration file in the code or configuration file writing window. In response to the user finishing writing the code or configuration file, the three-dimensional creation engine may obtain the user-written code or configuration file as virtual environment configuration information, and configure virtual light sources, virtual objects, and virtual cameras in the virtual environment based on the virtual environment configuration information. That is, the configuration operation may include but is not limited to the operation of opening the API and the input operation, and the input operation may include writing a code or configuration file. For example, in response to the user opening the API in the configuration interface, the three-dimensional creation engine may display a code or configuration file writing window corresponding to the API. Assuming that the content of the code or configuration file written by the user in the code or configuration file writing window is "the number of virtual light sources is 96 and the 96 virtual light sources are evenly arranged in a spherical array with a diameter of 8 decimeters", the three-dimensional creation engine may obtain the code or configuration file in response to finishing writing the code or configuration file, and based on finishing writing the code or configuration file, set a spherical array with a diameter of 8 decimeters in the virtual environment, and evenly set 96 virtual light sources in the spherical array.

[0118] It is understandable that, compared with the method of configuring parameters of virtual light sources, virtual objects, and virtual cameras one by one based on a configuration interface (such as a sub-interface), the method of writing code or configuration files is more efficient, but the corresponding professional requirements are higher. Compared with the method of writing code or configuration files, the method of configuring parameters of virtual light sources, virtual objects, and virtual cameras one by one based on a configuration interface (such as a sub-interface) is a little less efficient, but the user operation is simple and the professional requirements are low. You can choose which method to configure parameters according to actual needs, and the embodiments of the present application do not make specific restrictions.

[0119] Step S320: According to the configuration operation, obtain virtual environment configuration information, where the virtual environment configuration information includes configuration parameters of multiple virtual light sources, at least one virtual object, and multiple virtual cameras.

[0120] For the detailed description of step S320, please refer to step S310, which will not be repeated here.

[0121] Step S330: According to the virtual environment configuration information, multiple virtual light sources, at least one virtual object and multiple virtual cameras are set in the virtual environment, wherein the multiple virtual light sources are configured to be arranged in a spherical array and facing a target position, the target position is located within the spherical array, and at least one virtual object is located at the target position.

[0122] Step S340: Render at least one virtual object through multiple virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object.

[0123] Step S350: adjusting configuration parameters of the virtual environment after setting a plurality of virtual light sources, a plurality of virtual cameras and at least one virtual object according to the three-dimensional model data.

[0124] Based on step S310 to step S350, the present application can simulate a multi-light source multi-view shooting system (light stage) by setting multiple virtual light sources, at least one virtual object and multiple cameras in a virtual environment, and render at least one virtual object through multiple virtual cameras, so as to obtain the three-dimensional model data corresponding to at least one virtual object, so as to facilitate the subsequent evaluation of the configuration of the light stage based on the three-dimensional model data, and help to optimize the light stage configuration before the real light stage is built, reduce the difficulty and cost of the real light stage, and make the real light stage meet the user's needs. In addition, based on the three-dimensional model data generated by the simulated light stage, the performance of the software module of the light stage can be verified, so as to facilitate the parallel development of the hardware and software of the light stage. In addition, a configuration interface is provided for users to configure virtual light sources, virtual objects and virtual cameras, so that the virtual environment configuration can be more in line with user needs, so that different virtual light stages can be simulated according to different user needs, which is helpful to refer to the configuration of the virtual light stage and build a real light stage that meets user needs.

[0125] Please refer to the figure Fig.28 , Fig.28 The structure diagram of the virtual environment configuration parameter adjustment device provided by the embodiment of the present application is shown. The virtual environment configuration parameter adjustment device 100 may include: a light source simulation module 110, an object simulation module 120, a camera simulation module 130, an object rendering module 140 and a parameter adjustment module 150.

[0126] The light source simulation module 110 is used to set a plurality of virtual light sources in a virtual environment. The plurality of virtual light sources are configured to be arranged in a spherical array and face a target position, and the target position is located within the spherical array.

[0127] The object simulation module 120 is used to set at least one virtual object in the virtual environment, and the at least one virtual object is located at the target position.

[0128] The camera simulation module 130 is used to set a plurality of virtual cameras in the virtual environment.

[0129] In some embodiments, the camera simulation module 130 is further configured to set the multiple virtual cameras to surround the at least one virtual object in the virtual environment.

[0130] The object rendering module 140 is used to render the at least one virtual object through the multiple virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object.

[0131] The parameter adjustment module 150 is used to adjust the configuration parameters of the virtual environment after setting a plurality of virtual light sources, a plurality of virtual cameras and at least one virtual object according to the three-dimensional model data.

[0132] In some embodiments, the parameter adjustment module 150 is further configured to adjust the configuration parameters of the plurality of virtual light sources according to the three-dimensional model data corresponding to the at least one virtual object.

[0133] In some embodiments, the parameter adjustment module 150 is also used to evaluate the configuration parameters of the multiple virtual light sources based on the three-dimensional model data corresponding to the at least one virtual object, using a de-rendering method and / or a photometric stereo imaging method; and adjust the configuration parameters of the multiple virtual light sources based on the evaluation results.

[0134] In some embodiments, the parameter adjustment module 150 is further configured to adjust the configuration parameters of the multiple virtual cameras according to the three-dimensional model data corresponding to the at least one virtual object.

[0135] In some embodiments, the parameter adjustment module 150 is further used to evaluate the configuration parameters of the multiple virtual cameras using a neural radiation field method based on the three-dimensional model data corresponding to the at least one virtual object; and adjust the configuration parameters of the multiple virtual cameras based on the evaluation results.

[0136] In some embodiments, the device 100 for adjusting the configuration parameters of the virtual environment may further include a parameter configuration module, which is used to obtain virtual environment configuration information, wherein the virtual environment configuration information includes configuration parameters of the multiple virtual light sources, the at least one virtual object, and the multiple virtual cameras. The light source simulation module 110 is also used to set multiple virtual light sources in the virtual environment according to the virtual environment configuration information. The object simulation module 120 is also used to set the at least one virtual object in the virtual environment according to the virtual environment configuration information. The camera simulation module 130 is also used to set multiple virtual cameras in the virtual environment according to the virtual environment configuration information.

[0137] In some embodiments, the parameter configuration module is also used to detect configuration operations performed by the user in the configuration interface, where the configuration interface is used to configure the configuration parameters of the multiple virtual light sources, the at least one virtual object, and the multiple virtual cameras; and obtain the virtual environment configuration information based on the configuration operations.

[0138] Those skilled in the art can clearly understand that the above device provided in the embodiment of the present application can implement the method provided in the embodiment of the present application. The specific working process of the above-described device and module can refer to the corresponding process of the method in the embodiment of the present application, which will not be repeated here.

[0139] In the embodiments provided in the present application, the coupling, direct coupling or communication connection between the modules shown or discussed may be indirect coupling or communication coupling through some interfaces, devices or modules, and may be electrical, mechanical or other forms, and the embodiments of the present application do not impose specific limitations on this.

[0140] In addition, each functional module in the embodiment of the present application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or in the form of software functional modules.

[0141] See also Fig.29 , Fig.29 The schematic diagram of the structure of the electronic device provided by the embodiment of the present application is shown. The electronic device 200 may include a memory 210 and a processor 220. The memory 210 stores an application program. When the processor 220 calls the application program, the method provided by the embodiment of the present application is executed. The above-mentioned three-dimensional creation engine (software) is deployed in the electronic device 200.

[0142] The processor 220 may include one or more processing cores. The processor 220 uses various interfaces and lines to connect various parts of the entire electronic device 200, and is used to run or execute instructions, programs, code sets or instruction sets stored in the memory 210, and call to run or execute data stored in the memory 210, perform various functions of the electronic device 200 and process data.

[0143] The processor 220 can be implemented in at least one of the following hardware forms: digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor 220 can integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It is understandable that the above-mentioned modem may not be integrated into the processor 220, but may be implemented separately through a communication chip.

[0144] The memory 210 may include a random access memory (RAM) or a read-only memory (ROM). The memory 210 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 210 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the above-mentioned various method embodiments, etc. The data storage area may store data created by the electronic device 200 during use, etc.

[0145] The embodiment of the present application also provides a computer-readable storage medium, on which program code is stored, and when a processor calls the program code, the method provided in the embodiment of the present application is executed.

[0146] The computer readable storage medium can be an electronic storage such as a flash memory, an electrically erasable and editable read-only memory (Electrically-Erasable Programmable Read-Only Memory, referred to as EEPROM), an erasable and editable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), a hard disk or a ROM. In some embodiments, the computer readable storage medium includes a non-volatile computer readable medium (Non-Transitory Computer-Readable Storage Medium, referred to as Non-TCRSM). The computer readable storage medium has a storage space for the program code of any method step in the above method. These program codes can be read from one or more computer program products or written into one or more computer program products. The program code can be compressed in an appropriate form.

[0147] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for adjusting virtual environment configuration parameters, characterized in that: include: Setting a plurality of virtual light sources in a virtual environment, wherein the plurality of virtual light sources are arranged in a spherical array and face a target position, wherein the target position is located within the spherical array; Setting at least one virtual object in the virtual environment, wherein the at least one virtual object is located at the target position; Setting a plurality of virtual cameras in the virtual environment; Rendering the at least one virtual object through the multiple virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object; According to the three-dimensional model data, configuration parameters of the virtual environment after setting the multiple virtual light sources, the multiple virtual cameras and the at least one virtual object are adjusted.

2. The method according to claim 1, characterized in that The adjusting and setting of the configuration parameters of the virtual environment after the multiple virtual light sources, the multiple virtual cameras and the at least one virtual object includes: The configuration parameters of the plurality of virtual light sources are adjusted according to the three-dimensional model data corresponding to the at least one virtual object.

3. The method according to claim 2, characterized in that The adjusting the configuration parameters of the plurality of virtual light sources according to the three-dimensional model data corresponding to the at least one virtual object comprises: According to the three-dimensional model data corresponding to the at least one virtual object, the configuration parameters of the plurality of virtual light sources are evaluated by using a de-rendering method and / or a photometric stereo imaging method; According to the evaluation result, the configuration parameters of the multiple virtual light sources are adjusted.

4. The method according to claim 1, characterized in that: The adjusting and setting of the configuration parameters of the virtual environment after the multiple virtual light sources, the multiple virtual cameras and the at least one virtual object includes: The configuration parameters of the multiple virtual cameras are adjusted according to the three-dimensional model data corresponding to the at least one virtual object.

5. The method according to claim 4, characterized in that The adjusting the configuration parameters of the plurality of virtual cameras according to the three-dimensional model data corresponding to the at least one virtual object comprises: According to the three-dimensional model data corresponding to the at least one virtual object, the configuration parameters of the multiple virtual cameras are evaluated using a neural radiation field method; According to the evaluation results, the configuration parameters of the multiple virtual cameras are adjusted.

6. The method according to claim 1, characterized in that The step of setting a plurality of virtual cameras in the virtual environment comprises: The multiple virtual cameras are arranged in the virtual environment to surround the at least one virtual object.

7. The method according to any one of claims 1 to 6, characterized in that: Before setting a plurality of virtual light sources in the virtual environment, the method further comprises: Acquire virtual environment configuration information, where the virtual environment configuration information includes configuration parameters of the multiple virtual light sources, the at least one virtual object, and the multiple virtual cameras; The method of setting a plurality of virtual light sources in a virtual environment, setting at least one virtual object in the virtual environment, and setting a plurality of virtual cameras in the virtual environment includes: According to the virtual environment configuration information, a plurality of virtual light sources, the at least one virtual object and a plurality of virtual cameras are arranged in the virtual environment.

8. The method according to claim 7, characterized in that The obtaining of virtual environment configuration information includes: Detecting a configuration operation performed by a user on a configuration interface, wherein the configuration interface is used to configure configuration parameters of the multiple virtual light sources, the at least one virtual object, and the multiple virtual cameras; According to the configuration operation, the virtual environment configuration information is obtained.

9. The method according to claim 7, characterized in that: The configuration parameters of the multiple virtual light sources, the at least one virtual object and the multiple virtual cameras include at least one or a combination of several parameters including quantity, position, shape, direction, size and operation mode parameters.

10. The method according to claim 9, characterized in that The configuration parameters of the plurality of virtual light sources include operating mode parameters, and the operating mode parameters include a dynamic change mode and a lighting intensity and a color corresponding to the dynamic change mode.

11. A device for adjusting virtual environment configuration parameters, characterized in that: include: A light source simulation module, used for setting a plurality of virtual light sources in a virtual environment, wherein the plurality of virtual light sources are arranged in a spherical array and face a target position, wherein the target position is located within the spherical array; An object simulation module, configured to set at least one virtual object in the virtual environment, wherein the at least one virtual object is located at the target position; A camera simulation module, used for setting a plurality of virtual cameras in the virtual environment; An object rendering module, configured to render the at least one virtual object through the multiple virtual cameras to obtain three-dimensional model data corresponding to the at least one virtual object; The parameter adjustment module is used to adjust the configuration parameters of the virtual environment after setting the multiple virtual light sources, multiple virtual cameras and at least one virtual object according to the three-dimensional model data.

12. An electronic device, characterized in that: include: A memory and a processor, wherein an application is stored in the memory, and when the processor calls the application, the method according to any one of claims 1 to 10 is executed.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, and when the processor calls the program codes, the method according to any one of claims 1 to 10 is executed.

Images