Face modeling method and device and storage medium

Abstract

The invention relates to a face modeling method and device and a storage medium. The method comprises the following steps: determining a first human face three-dimensional model, and obtaining a human face image uploaded by a user; generating a target face three-dimensional grid model based on the first face three-dimensional model and the face image; and determining texture parameters and material parameters of the target face three-dimensional grid model based on the face image, and generating a second face three-dimensional model. According to the embodiment of the invention, the stylized three-dimensional face model can be automatically and efficiently generated based on the two-dimensional face image provided by the user, the user can efficiently and accurately obtain the expected three-dimensional face appearance without professional modeling experience or manual complex parameter adjustment in the process, and the user experience is improved. And the creation threshold of the high-quality three-dimensional face content is obviously reduced.

Description

Technical Field

[0001] This disclosure relates to the field of artificial intelligence technology, and in particular to a face modeling method, apparatus and storage medium. Background Technology

[0002] With the rapid development of artificial intelligence technology, the generation and editing of 3D faces has made significant progress in the field of digital content creation. 3D face models are widely used in animation and film production, virtual reality, game character creation, and human-computer interaction, and their quality and editability directly affect the final visual presentation. Currently, some 3D character creation platforms typically construct the user's desired facial image through parametric sliders, preset template combinations, and limited appearance editing options. While these methods lower the professional threshold for 3D modeling to some extent, they still suffer from cumbersome operation processes, limited creative expression, insufficient fine-tuning capabilities, and weak controllability of generated results, making it difficult for users to efficiently and accurately obtain a 3D face appearance that meets their expectations. Summary of the Invention

[0003] In view of this, the present disclosure proposes a face modeling method, apparatus and storage medium.

[0004] According to one aspect of this disclosure, a face modeling method is provided. The method includes:

[0005] Determine the first 3D model of the face and acquire the face image uploaded by the user;

[0006] A target face 3D mesh model is generated based on the first face 3D model and the face image;

[0007] Based on the face image, determine the texture parameters and material parameters of the target face 3D mesh model, and generate a second face 3D model.

[0008] In one possible implementation, a target face 3D mesh model is generated based on a first face 3D model and a face image, including:

[0009] Based on facial images, the three-dimensional geometric structure of the face is inferred, and the initial three-dimensional mesh model of the face is determined.

[0010] The coordinates of the 3D mesh points in the initial 3D face mesh model and the coordinates of the 3D mesh points in the first 3D face model are fused to generate the target 3D face mesh model, where the 3D mesh points are used to indicate the 3D shape of the face.

[0011] In one possible implementation, the coordinates of 3D mesh points in the initial 3D face mesh model and the coordinates of 3D mesh points in the first 3D face model are fused to generate the target 3D face mesh model, including:

[0012] Semantic matching is performed on the 3D mesh points in the initial face 3D mesh model and the first face 3D model to determine the correspondence between the 3D mesh points in the initial face 3D mesh model and the 3D mesh points in the first face 3D model;

[0013] Based on the correspondence, the 3D grid points in the initial face 3D mesh model and the corresponding 3D grid points in the first face 3D model are weighted and fused according to preset weights to obtain the coordinates of the target 3D grid points; among them, the preset weights are different for different face regions;

[0014] A 3D mesh model of the target face is generated based on the coordinates of the target 3D mesh points.

[0015] In one possible implementation, a target face 3D mesh model is generated based on the coordinates of the target 3D mesh points, including:

[0016] Based on the matching similarity between the unfolded result of the target 3D mesh point coordinates on the 2D plane and the face image, the target 3D mesh point coordinates are optimized and adjusted to obtain the target face 3D mesh model.

[0017] In one possible implementation, the texture and material parameters of the target face 3D mesh model are determined based on the face image to generate a second face 3D model, including:

[0018] The surface unfolding and texture mapping of the target face 3D mesh model are performed based on the face image. The face image is mapped onto the 3D mesh surface of the target face 3D mesh model to obtain the texture parameters of the target face 3D mesh model.

[0019] Based on the texture parameters and the lighting characteristics of the face image, determine the material parameters of the target face 3D mesh model;

[0020] Texture and material parameters are applied to the target face 3D mesh model to generate a second face 3D model.

[0021] In one possible implementation, the method further includes:

[0022] The display shows multiple candidate second face 3D models, which are multiple face 3D models with different appearances generated based on the same first face 3D model and face image;

[0023] In response to the user's selection of one of the candidate second face 3D models, the user-selected second face 3D model is determined as the final second face 3D model, and the user-selected second face 3D model is displayed.

[0024] In one possible implementation, the method further includes:

[0025] In response to the user's selection of a tag, the appearance parameters of the second face 3D model are adjusted to generate at least one third face 3D model. The appearance parameters include hairstyle parameters and / or clothing parameters that match the tag. The tag is used to indicate the user's desired appearance style.

[0026] In response to the user's selection of one of the three-dimensional models of a third face, the selected three-dimensional model of the third face is displayed.

[0027] In one possible implementation, in response to a user's selection of a label, the appearance parameters of the second 3D face model are adjusted to generate at least one third 3D face model, including:

[0028] In response to the user's selection of a tag, determine the element that matches the tag from the material library;

[0029] Based on appearance elements, adjust the appearance parameters of the second face 3D model to generate at least one third face 3D model.

[0030] According to another aspect of this disclosure, a face modeling apparatus is provided. The apparatus includes:

[0031] The first determining module is used to determine the first three-dimensional model of the face and to obtain the face image uploaded by the user.

[0032] The second determining module is used to generate a target face 3D mesh model based on the first face 3D model and the face image;

[0033] The third determining module is used to determine the texture parameters and material parameters of the target face 3D mesh model based on the face image, and generate a second face 3D model.

[0034] In one possible implementation, the second determining module is used for:

[0035] Based on facial images, the three-dimensional geometric structure of the face is inferred, and the initial three-dimensional mesh model of the face is determined.

[0036] The coordinates of the 3D mesh points in the initial 3D face mesh model and the coordinates of the 3D mesh points in the first 3D face model are fused to generate the target 3D face mesh model, where the 3D mesh points are used to indicate the 3D shape of the face.

[0037] In one possible implementation, the coordinates of 3D mesh points in the initial 3D face mesh model and the coordinates of 3D mesh points in the first 3D face model are fused to generate the target 3D face mesh model, including:

[0038] Semantic matching is performed on the 3D mesh points in the initial face 3D mesh model and the first face 3D model to determine the correspondence between the 3D mesh points in the initial face 3D mesh model and the 3D mesh points in the first face 3D model;

[0039] Based on the correspondence, the 3D grid points in the initial face 3D mesh model and the corresponding 3D grid points in the first face 3D model are weighted and fused according to preset weights to obtain the coordinates of the target 3D grid points; among them, the preset weights are different for different face regions;

[0040] A 3D mesh model of the target face is generated based on the coordinates of the target 3D mesh points.

[0041] In one possible implementation, a target face 3D mesh model is generated based on the coordinates of the target 3D mesh points, including:

[0042] Based on the matching similarity between the unfolded result of the target 3D mesh point coordinates on the 2D plane and the face image, the target 3D mesh point coordinates are optimized and adjusted to obtain the target face 3D mesh model.

[0043] In one possible implementation, the third determining module is used for:

[0044] The surface unfolding and texture mapping of the target face 3D mesh model are performed based on the face image. The face image is mapped onto the 3D mesh surface of the target face 3D mesh model to obtain the texture parameters of the target face 3D mesh model.

[0045] Based on the texture parameters and the lighting characteristics of the face image, determine the material parameters of the target face 3D mesh model;

[0046] Texture and material parameters are applied to the target face 3D mesh model to generate a second face 3D model.

[0047] In one possible implementation, the device further includes:

[0048] The first display module is used to display multiple candidate second face 3D models, which are multiple face 3D models with different appearances generated based on the same first face 3D model and face image;

[0049] The fourth determination module is used to respond to the user's selection operation for one of the candidate second face 3D models, determine the second face 3D model selected by the user as the final second face 3D model, and display the second face 3D model selected by the user.

[0050] In one possible implementation, the device further includes:

[0051] An adjustment module is used to adjust the appearance parameters of a second face 3D model in response to a user's selection of a label, and generate at least one third face 3D model. The appearance parameters include hairstyle parameters and / or clothing parameters that match the label, and the label is used to indicate the user's desired appearance style.

[0052] The second display module is used to display the third face 3D model selected by the user in response to the user's selection operation for one of the third face 3D models.

[0053] In one possible implementation, the module is adjusted for:

[0054] In response to the user's selection of a tag, determine the element that matches the tag from the material library;

[0055] Based on appearance elements, adjust the appearance parameters of the second face 3D model to generate at least one third face 3D model.

[0056] According to another aspect of this disclosure, a face modeling apparatus is provided, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0057] According to another aspect of this disclosure, a non-volatile computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the above-described method.

[0058] According to another aspect of this disclosure, a computer program product is provided, including a computer program or a non-volatile computer-readable storage medium carrying the computer program, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0059] According to embodiments of this disclosure, by determining a first 3D face model and acquiring a user-uploaded face image, a target 3D face mesh model is generated based on the first 3D face model and the face image. The texture and material parameters of the target 3D face mesh model are then determined based on the face image, generating a second 3D face model. This allows for the automatic and efficient generation of stylized 3D face models from user-provided 2D face images. The generated 3D face models are geometrically highly consistent with the face shape in the user image. Furthermore, the realism and visual expressiveness of the model can be significantly enhanced through refined texture and material processing. In this process, users do not need professional modeling experience or to manually adjust complex parameters to efficiently and accurately obtain the expected 3D face appearance, significantly lowering the barrier to creating high-quality 3D face content.

[0060] Other features and aspects of this disclosure will become clear from the following detailed description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0061] The accompanying drawings, which are included in and form part of this specification, illustrate exemplary embodiments, features, and aspects of this disclosure together with the specification and serve to explain the principles of this disclosure.

[0062] Figure 1 A schematic diagram of an application scenario according to an embodiment of this disclosure.

[0063] Figure 2 A flowchart illustrating a face modeling method according to an embodiment of the present disclosure is shown.

[0064] Figure 3 A schematic diagram of a first three-dimensional face model according to an embodiment of the present disclosure is shown.

[0065] Figure 4 A schematic diagram illustrating a user-uploaded facial image according to an embodiment of the present disclosure is shown.

[0066] Figure 5 A schematic diagram illustrating model fusion according to an embodiment of the present disclosure is shown.

[0067] Figure 6 A schematic diagram of a three-dimensional mesh model of a target face according to an embodiment of the present disclosure is shown.

[0068] Figure 7 A schematic diagram of a second three-dimensional face model according to an embodiment of the present disclosure is shown.

[0069] Figure 8 A schematic diagram of a label according to an embodiment of the present disclosure is shown.

[0070] Figure 9 A schematic diagram showing the determination of appearance elements according to an embodiment of the present disclosure is provided.

[0071] Figure 10 A schematic diagram illustrating the generation of a third-person face 3D model according to an embodiment of the present disclosure is shown.

[0072] Figure 11 A schematic diagram illustrating the generation of a third-person face 3D model according to an embodiment of the present disclosure is shown.

[0073] Figure 12 A schematic diagram illustrating the generation of a third-person face 3D model according to an embodiment of the present disclosure is shown.

[0074] Figure 13 A structural diagram of a face modeling apparatus according to an embodiment of the present disclosure is shown.

[0075] Figure 14 This is a block diagram illustrating an apparatus 1900 for face modeling according to an exemplary embodiment. Detailed Implementation

[0076] Various exemplary embodiments, features, and aspects of this disclosure will now be described in detail with reference to the accompanying drawings. The same reference numerals in the drawings denote elements that have the same or similar functions. Although various aspects of the embodiments are shown in the drawings, they are not necessarily drawn to scale unless specifically indicated otherwise.

[0077] As used herein, the terms “comprising,” “including,” “having,” or variations thereof are open-ended and include one or more of the stated features, integrals, elements, steps, components, or functions, but do not exclude the presence or addition of one or more other features, integrals, elements, steps, components, functions, or groups thereof.

[0078] When an element is referred to as “connected,” “coupled,” “responding,” or a variation thereof relative to another element, it may be directly connected, coupled, or responding to another element, or there may be an intermediate element present.

[0079] Although the terms first, second, third, etc., may be used herein to describe various elements / operations, these elements / operations should not be limited by these terms. These terms are only used to distinguish one element / operation from another. Therefore, without departing from the teachings of the inventive concept, a first element / operation in some embodiments may be referred to as a second element / operation in other embodiments.

[0080] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.

[0081] Furthermore, to better illustrate this disclosure, numerous specific details are set forth in the following detailed description. Those skilled in the art will understand that this disclosure can be practiced without certain specific details. In some instances, methods, means, components, and circuits well known to those skilled in the art have not been described in detail in order to highlight the main points of this disclosure.

[0082] It should be noted that the information (including but not limited to user device information, user personal information, etc.), data (including but not limited to data used for analysis, data stored, data displayed, etc.) and signals involved in this application are all authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant regions.

[0083] With the rapid development of artificial intelligence (AI) technology, the generation and editing of 3D faces has made significant progress in the field of digital content creation. 3D face models are widely used in animation and film production, virtual reality, game character creation, and human-computer interaction, and their quality and editability directly affect the final visual presentation. Currently, some 3D character creation platforms typically construct the user's desired facial image through parametric sliders, preset template combinations, and limited appearance editing options. While these methods lower the professional threshold for 3D modeling to some extent, they still suffer from cumbersome operation processes, limited creative expression, insufficient fine-tuning capabilities, and weak controllability of generated results, making it difficult for users to efficiently and accurately obtain a 3D face appearance that meets their expectations.

[0084] In view of this, this disclosure provides a face modeling method, apparatus, and storage medium. The method of this disclosure determines a first 3D face model and acquires a face image uploaded by a user. Based on the first 3D face model and the face image, a target 3D face mesh model is generated. Based on the face image, the texture and material parameters of the target 3D face mesh model are determined, and a second 3D face model is generated. This method can automatically and efficiently generate stylized 3D face models based on user-provided 2D face images. The generated 3D face models are highly consistent with the face shape in the user image in terms of geometric structure. Furthermore, the realism and visual expressiveness of the model can be significantly improved through refined texture and material processing. In this process, users do not need professional modeling experience or to manually adjust complex parameters to efficiently and accurately obtain the expected 3D face appearance, significantly lowering the barrier to creating high-quality 3D face content.

[0085] Figure 1 A schematic diagram illustrating an application scenario according to an embodiment of this disclosure. For example... Figure 1 As shown, the method of this disclosure can be used in a face modeling platform, for example, as an AI-assisted tool integrated into the modeling platform. Users can interact with the face modeling platform through terminal devices, etc.

[0086] First, a baseline 3D facial model can be determined (i.e., the aforementioned first 3D facial model, which can be determined by the user or generated by the platform). Subsequently, the user can upload one or more images containing faces through the platform (e.g., using an interactive interface provided by an AI-assisted tool). Based on this, the method of this disclosure embodiment can be used to automatically process the user input and generate a fused 3D facial model (i.e., the aforementioned second 3D facial model), thereby achieving the automated construction of a high-quality 3D facial image.

[0087] The AI-assisted tool in this embodiment is a tool that is easy for users to adjust, has a low learning cost, and allows for the mixing of multiple parameters to create a human figure, in order to help users, understand users, and quickly generate a 3D facial model.

[0088] This face modeling platform can be deployed on terminal devices or servers. The terminal devices involved in this disclosure can be any one or more of the following: mobile phones, foldable electronic devices, tablet computers, desktop computers, laptop computers, handheld computers, laptops, ultra-mobile personal computers (UMPCs), netbooks, cellular phones, personal digital assistants (PDAs), and in-vehicle devices. This application does not impose any special limitations on the specific type of terminal device; it can have wired or wireless communication capabilities.

[0089] The server disclosed herein can be located locally or in the cloud, and can be a physical device or a virtual device, such as a virtual machine or container. It possesses wireless communication capabilities, which can be configured within the server's chip (system) or other components. These wireless communication capabilities can be implemented through mobile communication technologies such as 2G / 3G / 4G / 5G, as well as Wi-Fi, Bluetooth, frequency modulation (FM), data radio, and satellite communication. Alternatively, communication can be achieved via a wired connection to enable interaction with other devices.

[0090] Figure 2 A flowchart illustrating a face modeling method according to an embodiment of this disclosure is shown. Figure 2 As shown, the method may include:

[0091] Step S201: Determine the first 3D face model and obtain the face image uploaded by the user.

[0092] The first three-dimensional face model can represent the baseline face model in the embodiments of this disclosure. The three-dimensional face model refers to a digital model that expresses the geometric structure and appearance features of a user's face in three-dimensional space. It may include facial parameters (such as the position, size, and shape parameters of facial features), three-dimensional geometric structure (such as topology and vertex position), applied material texture parameters (such as makeup texture), appearance (such as hairstyle), and other information.

[0093] The first 3D face model can be recommended from the preset template library. The preset template library can store a variety of different styles of 3D face templates. Based on user preferences, historical selection records, target scene requirements, etc., a basic model can be intelligently recommended to the user from the preset template library.

[0094] The first 3D facial model can also be customized by the user. For example, users can use parametric controls (such as sliders, switches, and option buttons) to personalize the facial skeleton, facial proportions, makeup style, skin texture, hairstyle, and clothing of the basic template—a process known as "face sculpting." Through these user interaction methods, a first 3D facial model that meets the user's needs can be generated.

[0095] Figure 3 A schematic diagram of a first three-dimensional face model according to an embodiment of the present disclosure is shown. Figure 3 As shown, users can adjust the face, makeup, clothing, hair, and other features of the 3D face model displayed in the top, bottom, left, and right sides of the interface using parametric controls. After completing the personalized editing, the user will obtain the first 3D face model shown in the center of the image.

[0096] A facial image can be two-dimensional image data captured or uploaded by a user. The image can include facial regions at any pose and angle, such as a frontal face or a side face, to facilitate subsequent processing.

[0097] Figure 4 A schematic diagram illustrating a user-uploaded facial image according to an embodiment of this disclosure is shown. Figure 4 As shown, users can upload facial images through the interactive interface provided by the AI-assisted tools shown in the figure (such as the module labeled "NOVA AI" in the figure).

[0098] This allows for the innovative integration of user multimodal input with the current 3D model state, generating precise operation commands that can directly drive the platform, thus achieving efficient and accurate modeling transformation.

[0099] Step S202: Generate a target face 3D mesh model based on the first face 3D model and the face image.

[0100] Figure 5 A schematic diagram illustrating model fusion according to an embodiment of this disclosure is shown. Figure 5 As shown, after a user uploads a facial image through the AI-assisted tool "NOVA AI," the trained model integrated into the tool (e.g., using SF3D or related technologies) can be used to fuse the first 3D facial model and the facial image to generate a target 3D facial mesh model. During the model fusion process, the generation progress can be visualized in real-time within the "NOVA AI" module, allowing users to easily understand the processing status.

[0101] Figure 6 A schematic diagram of a three-dimensional mesh model of a target face according to an embodiment of the present disclosure is shown. Figure 6As shown, the 3D mesh model of the target face can be displayed in the "NOVA AI" module. This target face 3D mesh model is a grayscale model, meaning it represents the 3D geometry of the face based on 3D mesh points and their spatial coordinates, but does not include texture, material, or lighting information.

[0102] In one possible implementation, step S202 includes:

[0103] Based on the facial image, the three-dimensional geometric structure of the face is inferred, and an initial three-dimensional mesh model of the face is determined. The coordinates of the three-dimensional mesh points in the initial three-dimensional mesh model of the face are fused with the coordinates of the three-dimensional mesh points in the first three-dimensional facial model to generate a target three-dimensional mesh model of the face.

[0104] One approach is to utilize existing 3D face reconstruction techniques to process face images, inferring the 3D geometric structure of the face and obtaining an initial 3D face mesh model. For example, a fitting algorithm based on a 3D Morphable Model (3DMM) can be used to infer the 3D geometric structure of the face based on key points extracted from the face image, generating an initial 3D face mesh model containing the coordinates of 3D mesh points.

[0105] Three-dimensional mesh points can be used to indicate the three-dimensional shape of a human face, and their coordinates can represent the spatial distribution of facial structures. For example, three-dimensional mesh points can be used to describe the three-dimensional configuration of key facial areas such as the bridge of the nose, the tip of the nose, the contour of the eye sockets, the cheekbones, and the jawline.

[0106] The 3D mesh points in the initial 3D face mesh model and the 3D mesh points in the first 3D face model can be processed by coordinate normalization and pose alignment to place them in a unified coordinate system, so as to ensure that the subsequent weighted fusion processing is carried out under the same spatial reference.

[0107] By fusing the coordinates of the 3D mesh points corresponding to the initial 3D face model and the first 3D face model, the personalized geometric features reflected by the face image in the first 3D face model can be preserved. Furthermore, the overall structure of the initial 3D face mesh model can be used as a shape constraint, ensuring stability and balance between details and overall contour in the fused model. This allows for the generation of a target 3D face mesh model that better matches the user's desired appearance.

[0108] In one possible implementation, the coordinates of 3D mesh points in the initial 3D face mesh model and the coordinates of 3D mesh points in the first 3D face model are fused to generate the target 3D face mesh model, including:

[0109] Semantic matching is performed on the 3D mesh points in the initial face 3D mesh model and the first face 3D model to determine the correspondence between the 3D mesh points in the initial face 3D mesh model and the 3D mesh points in the first face 3D model. Based on the correspondence, the 3D mesh points in the initial face 3D mesh model and the corresponding 3D mesh points in the first face 3D model are weighted and fused according to preset weights to obtain the coordinates of the target 3D mesh points. Based on the coordinates of the target 3D mesh points, the target face 3D mesh model is generated.

[0110] Specifically, the topological structure and feature point distribution of the two sets of 3D mesh points can be aligned based on a preset facial semantic structure (such as semantic regions like the corners of the eyes, brow peaks, nose tip, nose wings, corners of the mouth, and jawline). For example, existing related technologies (such as facial key point detection technology) can be used to determine the semantic correspondence of each 3D mesh point in the initial model in the first 3D face model.

[0111] After establishing semantic correspondence, the corresponding 3D mesh points can be weighted and fused based on preset weights to obtain the target 3D mesh point coordinates. For example, for the i-th corresponding 3D mesh point, its target 3D mesh point coordinates can be obtained by the following formula: P_i,target = w_i,1×P_i,1 + w_i,2×P_i,2. Where, P_i,target is the coordinate of the i-th target 3D mesh point, P_i,1 and P_i,2 are the coordinates of the i-th corresponding 3D mesh points in the initial face 3D mesh model and the first face 3D model, respectively, and w_i,1 and w_i,2 are the preset weights corresponding to the semantic regions to which P_i,1 and P_i,2 belong.

[0112] The preset weights for different facial regions can vary. Different weights can be preset based on the sensitivity and structural stability of different facial regions. For example, in regions with significant individual features such as the eyes, nose tip, and mouth, higher weights can be assigned to the corresponding 3D mesh points in the initial 3D facial mesh model (e.g., w_i,1 set to 70%, w_i,2 set to 30%) to preserve the realistic features of the facial image. In structural regions such as the cheeks, jawline, and forehead, the weights of the 3D mesh points in the first 3D facial model can be increased (e.g., w_i,1 set to 30%, w_i,2 set to 70%) to ensure the smoothness and stability of the overall facial contour.

[0113] After obtaining the coordinates of all target 3D mesh points, the target 3D mesh points can be connected based on preset mesh connection relationships (such as triangular facet index table, vertex connection relationship, etc.) to form a target face 3D mesh model.

[0114] To further match the facial features reflected in the face image in terms of 3D shape, the generated 3D mesh model of the target face is generated based on the coordinates of the target 3D mesh points, including:

[0115] Based on the matching similarity between the unfolded result of the target 3D mesh point coordinates on the 2D plane and the face image, the target 3D mesh point coordinates are optimized and adjusted to obtain the target face 3D mesh model.

[0116] Specifically, a 3D face mesh model obtained based on the coordinates of the target 3D mesh points can be mapped to a 2D plane in the pixel coordinate system of the face image using existing geometric unfolding algorithms, resulting in corresponding 2D projected coordinates. These 2D projected coordinates are then compared with the face image to determine the matching similarity. The matching similarity can be evaluated using existing techniques, such as reprojection error, which represents the difference in contours, texture gradients, etc., between the 3D mesh points mapped to 2D and the face image.

[0117] When the similarity is less than a preset threshold, the coordinates of the target 3D mesh points can be optimized and adjusted. Optimization methods may include techniques such as local deformation, key point position correction, or overall geometric fine-tuning to reduce the error with the face image and improve the fit between the 3D mesh model and the face image.

[0118] When the matching similarity is not less than a preset threshold, it means that the matching between the target 3D mesh model and the face image has achieved the expected effect, and the face 3D mesh model obtained based on the coordinates of the target 3D mesh points can be used as the target face 3D mesh model.

[0119] Step S203: Determine the texture parameters and material parameters of the target face 3D mesh model based on the face image, and generate a second face 3D model.

[0120] Texture parameters can be used to indicate the color, texture details, and lighting characteristics of the face surface. For example, they can include texture coordinates and texture color. Material parameters can be used to indicate the optical properties of the face surface, material reflection, and lighting interaction effects. For example, they can include parameters such as diffuse reflectance, specular reflectance, specular intensity, and roughness.

[0121] According to embodiments of this disclosure, by determining a first 3D face model and acquiring a user-uploaded face image, a target 3D face mesh model is generated based on the first 3D face model and the face image. The texture and material parameters of the target 3D face mesh model are then determined based on the face image, generating a second 3D face model. This allows for the automatic and efficient generation of stylized 3D face models from user-provided 2D face images. The generated 3D face models are geometrically highly consistent with the face shape in the user image. Furthermore, the realism and visual expressiveness of the model can be significantly enhanced through refined texture and material processing. In this process, users do not need professional modeling experience or to manually adjust complex parameters to efficiently and accurately obtain the expected 3D face appearance, significantly lowering the barrier to creating high-quality 3D face content.

[0122] In one possible implementation, step S203 includes:

[0123] The surface of the target face 3D mesh model is unfolded and texture mapped based on the face image. The face image is mapped onto the 3D mesh surface of the target face 3D mesh model to obtain the texture parameters of the target face 3D mesh model. Based on the texture parameters and the lighting characteristics of the face image, the material parameters of the target face 3D mesh model are determined. The texture parameters and material parameters are applied to the target face 3D mesh model to generate a second face 3D model.

[0124] This process can employ existing technologies (such as geometric unfolding algorithms) to map the coordinates of 3D mesh points in the target face's 3D mesh model onto a 2D plane, generating a corresponding UV unfolded map (i.e., the unfolded result including 2D texture coordinates). Based on this, according to the correspondence between the 2D texture coordinates and the 2D coordinates in the face image (e.g., determined based on the aforementioned semantic correspondence), color sampling can be performed on corresponding pixels in the face image, and the resulting pixel color can be used as the texture color for the corresponding 2D texture coordinates. Thus, the texture coordinates and texture colors corresponding to each mesh point in the target face's 3D mesh model can be obtained as texture parameters.

[0125] Furthermore, the lighting characteristics of the facial image can be analyzed. For example, existing lighting estimation algorithms can be used to determine information such as the overall and local lighting intensity, brightness gradient distribution, surface reflection characteristics, and shadow areas of the facial image. Combined with the color distribution obtained after texture mapping, the reflection properties of the target facial 3D mesh model's surface can be further determined, such as determining diffuse reflection coefficient, specular reflection coefficient, roughness parameters, and normal detail enhancement parameters, among other material parameters. This allows the generated second 3D facial model to present a more realistic visual effect under different lighting conditions.

[0126] Figure 7A schematic diagram of a second three-dimensional face model according to an embodiment of the present disclosure is shown. Figure 7 As shown, a second 3D face model can be generated and displayed based on the target face 3D mesh model displayed in the "NOVA AI" module by determining the texture parameters and material parameters of the target face 3D mesh model.

[0127] In one possible implementation, the method further includes:

[0128] Display multiple candidate second face 3D models; in response to the user's selection operation for one of the candidate second face 3D models, determine the second face 3D model selected by the user as the final second face 3D model, and display the second face 3D model selected by the user.

[0129] The multiple candidate second-face 3D models can be generated from the same first-face 3D model and face image, each with a different appearance. Different appearances can refer to differences in any one or more of the following: 3D geometric structure, texture parameters, and material parameters, to provide users with second-face 3D models of varying stylization and detail optimization. For example, by controlling the weights of corresponding 3D mesh points in the initial face 3D mesh model and the weights of 3D mesh points in the first face 3D model, second-face 3D models with different stylization levels can be obtained; or by adjusting the texture or material parameters, second-face 3D models with different texture or material rendering effects can be obtained.

[0130] This allows for the automatic generation of multiple candidate 3D face models with differentiated features based on the same input, using AI-assisted tools. These models are then visualized and presented to the user, enabling intuitive comparison between the different candidate models. Users can select one candidate model as their final second 3D face model based on their personal preference. This allows for direct user participation and style control without increasing modeling complexity, improving the controllability of the 3D face model generation process and enhancing the user experience.

[0131] In one possible implementation, the method further includes:

[0132] In response to the user's selection of a label, adjust the appearance parameters of the second face 3D model to generate at least one third face 3D model; in response to the user's selection of one of the third face 3D models, display the third face 3D model selected by the user.

[0133] The labels can be pre-configured labels, or labels determined by the AI-assisted tool based on semantic matching and other methods after the user inputs relevant appearance requirements through the interaction interface of the aforementioned AI-assisted tool. These labels are used to indicate the user's desired appearance style. Figure 8 A schematic diagram of a label according to an embodiment of the present disclosure is shown. Taking the generation of a hairstyle appearance as an example, such as... Figure 8 Below the second 3D face model shown, users can click the "Match Character Hairstyle to Avatar" button control to expand and display labels such as "Samurai", "Beauty", "Cyber", and "Maid".

[0134] Therefore, the appearance of a 3D face model can be quickly stylized according to the user's style preferences.

[0135] Appearance parameters may include hairstyle parameters and / or clothing parameters that match the tags. Hairstyle parameters can be used to indicate the hairstyle appearance of the target 3D model, including but not limited to hairstyle type, hair length, hair color, hair volume, curl, and bangs shape. In one possible implementation, the hairstyle appearance of the 3D face model can be determined based on a combination of preset hair sections (e.g., top hair sections, side hair sections, back hair sections, bangs hair sections, etc.) and corresponding material and texture settings.

[0136] Clothing parameters can be used to indicate the appearance of clothing on the upper body of a 3D facial model, including but not limited to clothing type, style, fabric texture, color scheme, and the position and selection of decorative elements. The clothing appearance of a 3D facial model can be determined by calling preset clothing model fragments.

[0137] In one possible implementation, in response to a user's selection of a label, the appearance parameters of the second 3D face model are adjusted to generate at least one third 3D face model, including:

[0138] In response to the user's selection of a tag, the system determines the appearance elements that match the tag from the material library; based on the appearance elements, it adjusts the appearance parameters of the second face 3D model to generate at least one third face 3D model.

[0139] The resource library can store various appearance elements of different styles, including preset hair extensions for determining hairstyles and preset clothing model fragments for determining clothing appearances. Based on the correspondence between user-selected tags and appearance elements in the resource library, it can determine hair extensions and / or clothing model fragments that match the style indicated by the tags.

[0140] Figure 9 This diagram illustrates the determination of appearance elements according to an embodiment of the present disclosure. Taking the generation of hairstyle appearance as an example, when the user selects, such as... Figure 9 After the "beautiful woman" tag shown, AI-assisted tools can be used to match the corresponding appearance elements from the material library.

[0141] To provide users with a wider range of appearance options, a single tag can be associated with one or more appearance elements. For example, the same "beautiful woman" tag can correspond to multiple hair extension combinations or multiple clothing model fragments. These candidate hair extensions or clothing model fragments can be combined and matched individually, and the appearance parameters of the second 3D face model can be adjusted accordingly to generate multiple third 3D face models as alternative options with different styles. Figure 10 , Figure 11 and Figure 12 Schematic diagrams illustrating the generation of a third-party face 3D model according to embodiments of this disclosure are shown. Figure 10 As shown, multiple alternative 3D models of third-party faces can be generated based on the user's selected "beautiful woman" tag. Figure 11 As shown, users can select a 3D face model for the final application. Figure 12 As shown, after a user selects a 3D face model from multiple third-party 3D face models, the selected 3D face model can be displayed in the center of the interface.

[0142] Therefore, it is possible to automatically generate 3D face models with various appearance styles based on the user's style preferences without requiring the user to perform complex manual modeling, which significantly improves the efficiency of 3D content creation and enhances the user's personalized experience.

[0143] In one possible implementation, the appearance features of the second 3D face model can be combined to automatically recommend suitable hairstyle and / or clothing parameters, and the second 3D face model can be adjusted to generate at least one third 3D face model. Users can then select one of these third 3D face models, displaying the selected model.

[0144] The AI-assisted tool can analyze the facial geometry (such as facial contour, jaw width, and cheekbone height), proportional features (such as the distance between facial features, forehead height, and nose length ratio) and skin color distribution of the second-person 3D model. Based on preset style matching rules or a style recommendation model trained on sample data, it can automatically select hair combination combinations or clothing model fragments that match the style of the face from the material library. For example, if the second-person 3D model is detected to have a long face, it can match hairstyles with strong layers or bangs to visually shorten the vertical proportion; if the second-person 3D model is detected to have a light skin tone or a cool color temperature, it can match clothing with low color contrast and a soft style.

[0145] Therefore, it is possible to automatically generate candidate schemes for third-person 3D models that are coordinated with facial features and have a unified overall style without relying on additional user input.

[0146] Figure 13A structural diagram of a face modeling apparatus according to an embodiment of the present disclosure is shown. Figure 13 As shown, the device may include:

[0147] The first determining module 1301 is used to determine the first three-dimensional face model and obtain the face image uploaded by the user;

[0148] The second determining module 1302 is used to generate a target face 3D mesh model based on the first face 3D model and the face image;

[0149] The third determining module 1303 is used to determine the texture parameters and material parameters of the target face 3D mesh model based on the face image, and generate a second face 3D model.

[0150] In one possible implementation, the second determining module 1302 is used for:

[0151] Based on facial images, the three-dimensional geometric structure of the face is inferred, and the initial three-dimensional mesh model of the face is determined.

[0152] The coordinates of the 3D mesh points in the initial 3D face mesh model and the coordinates of the 3D mesh points in the first 3D face model are fused to generate the target 3D face mesh model, where the 3D mesh points are used to indicate the 3D shape of the face.

[0153] In one possible implementation, the coordinates of 3D mesh points in the initial 3D face mesh model and the coordinates of 3D mesh points in the first 3D face model are fused to generate the target 3D face mesh model, including:

[0154] Semantic matching is performed on the 3D mesh points in the initial face 3D mesh model and the first face 3D model to determine the correspondence between the 3D mesh points in the initial face 3D mesh model and the 3D mesh points in the first face 3D model;

[0155] Based on the correspondence, the 3D grid points in the initial face 3D mesh model and the corresponding 3D grid points in the first face 3D model are weighted and fused according to preset weights to obtain the coordinates of the target 3D grid points; among them, the preset weights are different for different face regions;

[0156] A 3D mesh model of the target face is generated based on the coordinates of the target 3D mesh points.

[0157] In one possible implementation, a target face 3D mesh model is generated based on the coordinates of the target 3D mesh points, including:

[0158] Based on the matching similarity between the unfolded result of the target 3D mesh point coordinates on the 2D plane and the face image, the target 3D mesh point coordinates are optimized and adjusted to obtain the target face 3D mesh model.

[0159] In one possible implementation, the third determining module 1303 is used for:

[0160] The surface unfolding and texture mapping of the target face 3D mesh model are performed based on the face image. The face image is mapped onto the 3D mesh surface of the target face 3D mesh model to obtain the texture parameters of the target face 3D mesh model.

[0161] Based on the texture parameters and the lighting characteristics of the face image, determine the material parameters of the target face 3D mesh model;

[0162] Texture and material parameters are applied to the target face 3D mesh model to generate a second face 3D model.

[0163] In one possible implementation, the device further includes:

[0164] The first display module is used to display multiple candidate second face 3D models, which are multiple face 3D models with different appearances generated based on the same first face 3D model and face image;

[0165] The fourth determination module is used to respond to the user's selection operation for one of the candidate second face 3D models, determine the second face 3D model selected by the user as the final second face 3D model, and display the second face 3D model selected by the user.

[0166] In one possible implementation, the device further includes:

[0167] An adjustment module is used to adjust the appearance parameters of a second face 3D model in response to a user's selection of a label, and generate at least one third face 3D model. The appearance parameters include hairstyle parameters and / or clothing parameters that match the label, and the label is used to indicate the user's desired appearance style.

[0168] The second display module is used to display the third face 3D model selected by the user in response to the user's selection operation for one of the third face 3D models.

[0169] In one possible implementation, the module is adjusted for:

[0170] In response to the user's selection of a tag, determine the element that matches the tag from the material library;

[0171] Based on appearance elements, adjust the appearance parameters of the second face 3D model to generate at least one third face 3D model.

[0172] According to embodiments of this disclosure, by determining a first 3D face model and acquiring a user-uploaded face image, a target 3D face mesh model is generated based on the first 3D face model and the face image. The texture and material parameters of the target 3D face mesh model are then determined based on the face image, generating a second 3D face model. This allows for the automatic and efficient generation of stylized 3D face models from user-provided 2D face images. The generated 3D face models are geometrically highly consistent with the face shape in the user image. Furthermore, the realism and visual expressiveness of the model can be significantly enhanced through refined texture and material processing. In this process, users do not need professional modeling experience or to manually adjust complex parameters to efficiently and accurately obtain the expected 3D face appearance, significantly lowering the barrier to creating high-quality 3D face content.

[0173] In some embodiments, the functions or modules of the apparatus provided in this disclosure can be used to perform the methods described in the above method embodiments. The specific implementation can be referred to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0174] This disclosure also provides a face modeling apparatus, including a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above method.

[0175] This disclosure also provides a non-volatile computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the above-described method.

[0176] This disclosure also provides a computer program product, including a computer program or a non-volatile computer-readable storage medium carrying the computer program, wherein the computer program, when executed by a processor, implements the steps of the above method.

[0177] Figure 14 This is a block diagram illustrating an apparatus 1900 for face modeling according to an exemplary embodiment. For example, apparatus 1900 may be provided as a server or terminal device. (Refer to...) Figure 14 The apparatus 1900 includes a processing component 1922, which further includes one or more processors, and memory resources represented by memory 1932 for storing instructions, such as application programs, that can be executed by the processing component 1922. The application programs stored in memory 1932 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 1922 is configured to execute instructions to perform the methods described above.

[0178] Device 1900 may also include a power supply component 1926 configured to perform power management of device 1900, a wired or wireless network interface 1950 configured to connect device 1900 to a network, and an input / output interface 1958 (I / O interface). Device 1900 can operate on an operating system, such as Windows Server, stored in memory 1932. TM macOS X TM Unix TM Linux TM FreeBSD TM Or similar.

[0179] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions that can be executed by a processing component 1922 of the device 1900 to perform the above-described method.

[0180] Computer-readable storage media can be tangible devices capable of holding and storing programs / instructions used by instruction execution devices. Computer-readable storage media can be, for example—but not limited to—electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination of the foregoing. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0181] The computer program (or computer-readable program instructions) described herein can be downloaded from a computer-readable storage medium to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage medium in the respective computing / processing device.

[0182] The computer program (or computer program instructions) used to perform the operations of this disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions to implement various aspects of this disclosure.

[0183] Various aspects of this disclosure are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0184] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0185] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0186] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0187] The various embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, nor are they limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A face modeling method, characterized in that, The method includes: Determine the first 3D model of the face and acquire the face image uploaded by the user; A target face 3D mesh model is generated based on the first face 3D model and the face image; Based on the face image, the texture parameters and material parameters of the target face 3D mesh model are determined, and a second face 3D model is generated.

2. The method according to claim 1, characterized in that, The step of generating a target face 3D mesh model based on the first face 3D model and the face image includes: Based on the face image, the three-dimensional geometric structure of the face is inferred, and an initial three-dimensional face mesh model is determined; The coordinates of the three-dimensional grid points in the initial three-dimensional face mesh model and the coordinates of the three-dimensional grid points in the first three-dimensional face model are fused to generate the target three-dimensional face mesh model, wherein the three-dimensional grid points are used to indicate the three-dimensional shape of the face.

3. The method according to claim 2, characterized in that, The step of fusing the coordinates of the 3D mesh points in the initial 3D face mesh model with the coordinates of the 3D mesh points in the first 3D face model to generate the target 3D face mesh model includes: Semantic matching is performed between the initial 3D face mesh model and the 3D mesh points in the first 3D face model to determine the correspondence between the 3D mesh points in the initial 3D face mesh model and the 3D mesh points in the first 3D face model. Based on the correspondence, the three-dimensional grid points in the initial three-dimensional face mesh model and the corresponding three-dimensional grid points in the first three-dimensional face model are weighted and fused according to preset weights to obtain the coordinates of the target three-dimensional grid points; wherein, the preset weights are different for different face regions; Based on the coordinates of the target 3D grid points, a 3D grid model of the target face is generated.

4. The method according to claim 3, characterized in that, The step of generating the target face 3D mesh model based on the target 3D mesh point coordinates includes: Based on the matching similarity between the unfolded result of the target 3D grid point coordinates on the 2D plane and the face image, the target 3D grid point coordinates are optimized and adjusted to obtain the target face 3D grid model.

5. The method according to claim 1, characterized in that, The step of determining the texture parameters and material parameters of the target face 3D mesh model based on the face image to generate a second face 3D model includes: Based on the face image, the surface of the target face 3D mesh model is unfolded and texture mapped, and the face image is mapped onto the 3D mesh surface of the target face 3D mesh model to obtain the texture parameters of the target face 3D mesh model; Based on the texture parameters and the lighting characteristics of the face image, the material parameters of the target face 3D mesh model are determined; The texture parameters and material parameters are applied to the target face 3D mesh model to generate a second face 3D model.

6. The method according to any one of claims 1-5, characterized in that, The method further includes: The system displays multiple candidate second face 3D models, which are multiple face 3D models with different appearances generated based on the same first face 3D model and face image. In response to the user's selection of one of the candidate second face 3D models, the user-selected second face 3D model is determined as the final second face 3D model, and the user-selected second face 3D model is displayed.

7. The method according to claim 1, characterized in that, The method further includes: In response to a user's selection of a label, the appearance parameters of the second 3D face model are adjusted to generate at least one third 3D face model. The appearance parameters include hairstyle parameters and / or clothing parameters that match the label, which is used to indicate the user's desired appearance style. In response to the user's selection of one of the three-dimensional models of the third face, the selected three-dimensional model of the third face is displayed.

8. The method according to claim 7, characterized in that, The step of adjusting the appearance parameters of the second 3D face model in response to the user's selection of a label, and generating at least one third 3D face model, includes: In response to the user's selection of a tag, determine the appearance element that matches the tag from the material library; Based on the appearance elements, adjust the appearance parameters of the second face 3D model to generate at least one third face 3D model.

9. A face modeling device, characterized in that, The device includes: The first determining module is used to determine the first three-dimensional model of the face and to obtain the face image uploaded by the user. The second determining module is used to generate a target face 3D mesh model based on the first face 3D model and the face image; The third determining module is used to determine the texture parameters and material parameters of the target face 3D mesh model based on the face image, and generate a second face 3D model.

10. A face modeling device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 8.

11. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

12. A computer program product comprising a computer program, or a non-volatile computer-readable storage medium carrying a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

Images