Search

CN-121998736-A - Self-adaptive fitting display recommendation method and system based on 3D clothing

CN121998736ACN 121998736 ACN121998736 ACN 121998736ACN-121998736-A

Abstract

The application provides a self-adaptive fitting display recommendation method and system based on 3D clothing, and relates to the technical field of clothing data processing. According to the technical scheme provided by the application, the three-dimensional basic grid model is subjected to interpolation deformation calculation through the target displacement vector, the deformation constraint weight and the expansion coefficient adjustment value to generate the target clothing model adapting to the target user, so that the clothing can truly fit with the body shape characteristics of the user. The dynamic try-on display picture is generated according to the cloth physical attribute parameters, and the pressure distribution thermodynamic diagram is generated based on the topological distance between the inner surface of the target clothing model and the outer surface of the three-dimensional human body model, so that the three-dimensional form, the fold distribution and the pressure relation of the clothing can be comprehensively presented, and the omnibearing real try-on experience can be provided for different users, and the try-on effect is improved.

Inventors

  • WANG JINGWU

Assignees

  • 山东圣梵尼服饰股份有限公司

Dates

Publication Date
20260508
Application Date
20260120

Claims (10)

  1. 1. A 3D garment-based self-adaptive try-on display recommendation method, the method comprising: acquiring body data of a target user, constructing a three-dimensional human body model, and carrying out semantic region segmentation on the three-dimensional human body model to obtain a plurality of body semantic regions and corresponding local curvature characteristics respectively; responding to a clothing selection instruction of the target user, calling a three-dimensional basic grid model of the target clothing, and calculating a target displacement vector corresponding to a characteristic control point preset on the three-dimensional basic grid model; Identifying grid vertexes corresponding to each body semantic region in the three-dimensional basic grid model, determining deformation constraint weights of the grid vertexes according to the types of the body semantic regions, and calculating expansion coefficient adjustment values of the grid vertexes in the normal direction according to the local curvature characteristics; performing interpolation deformation calculation on the three-dimensional basic grid model according to the target displacement vector, the deformation constraint weight and the expansion coefficient adjustment value to generate a target clothing model adapting to the target user; loading the target clothing model to the three-dimensional human body model, and generating a dynamic try-on display picture according to the cloth physical attribute parameters corresponding to the target clothing; Generating a pressure distribution thermodynamic diagram according to the topological distance between the inner surface of the target clothing model and the outer surface of the three-dimensional human body model, and generating clothing recommendation information based on the pressure distribution thermodynamic diagram and the dynamic try-on display picture.
  2. 2. The method according to claim 1, wherein the calculating the target displacement vector corresponding to the feature control point preset on the three-dimensional basic grid model includes: acquiring a standard human body model matched with the three-dimensional basic grid model, and extracting a first key point set on the standard human body model and a second key point set on the three-dimensional human body model; Calculating a local affine transformation matrix from the standard human body model to the three-dimensional human body model according to the corresponding relation between the first key point set and the second key point set; And determining the projection positions of the preset characteristic control points on the three-dimensional basic grid model on the standard human body model, and calculating coordinate differences of the characteristic control points mapped to the positions corresponding to the three-dimensional human body model according to the local affine transformation matrix corresponding to the projection positions to obtain a target displacement vector.
  3. 3. The method of claim 1, wherein the body semantic region comprises a rigid support region and a flexible overhang region, the determining deformation constraint weights for each of the mesh vertices based on a type of the body semantic region comprising: Acquiring a preset region weight configuration table and judging a body semantic region to which the grid vertex belongs; If the grid vertexes are positioned in the rigid supporting area, setting deformation constraint weights of the grid vertexes as first weight values; And if the grid vertex is positioned in the flexible suspension area, setting the deformation constraint weight of the grid vertex as a second weight value, wherein the second weight value is smaller than the first weight value.
  4. 4. A method according to claim 3, characterized in that the method further comprises: And if the grid vertex is positioned in the transition region between the rigid support region and the flexible suspension region, performing smoothing on the second weight value and the first weight value, and setting the deformation constraint weight of the grid vertex as a third weight value.
  5. 5. The method of claim 1, wherein said calculating an adjustment value of the expansion coefficient of each of the mesh vertices in the normal direction based on the local curvature features comprises: acquiring local curvature characteristics of a body semantic region corresponding to the grid vertexes, and calculating curvature values of the local curvature characteristics; And determining the geometric convex-concave characteristics of the area where the grid vertex is positioned according to the curvature value, and determining an adjustment vector corresponding to the curvature value from a preset expansion coefficient adjustment range to obtain an expansion coefficient adjustment value of the grid vertex in the normal direction.
  6. 6. The method of claim 5, wherein determining the geometric convex-concave characteristic of the area where the grid vertex is located according to the curvature value, determining an adjustment vector corresponding to the curvature value from a preset expansion coefficient adjustment range, and obtaining the expansion coefficient adjustment value of the grid vertex in the normal direction includes: If the curvature value is larger than a preset bulge threshold value, determining a first adjustment vector corresponding to the curvature value from a preset expansion coefficient adjustment range, and increasing the expansion coefficient adjustment value of the grid vertex in the normal direction according to the first adjustment vector; If the curvature value is smaller than a preset concave threshold value, determining a second adjustment vector corresponding to the curvature value from a preset expansion coefficient adjustment range, and reducing the expansion coefficient adjustment value of the grid vertex in the normal direction according to the second adjustment vector; And if the curvature value is smaller than or equal to the convex threshold value and larger than or equal to the concave threshold value, calculating an expansion coefficient adjustment value of the grid vertex by linear interpolation according to the difference value of the curvature value and the preset reference curvature.
  7. 7. The method of claim 1, wherein the generating garment recommendation information based on the pressure distribution thermodynamic diagram and the dynamic try-in display comprises: Acquiring a scene demand label of the target user, and determining a pressure threshold of the target user according to a scene type corresponding to the scene demand label; Counting the area ratio of the overpressure region exceeding a pressure threshold in the pressure distribution thermodynamic diagram, and calculating a comprehensive matching degree score by combining the frequency domain characteristics of the clothing folds in the dynamic try-on display picture; if the comprehensive matching degree score is higher than or equal to a preset score standard, generating recommended purchase information; and if the comprehensive matching degree score is lower than a preset score standard, generating size adjustment suggestion information or style adjustment suggestion information according to the position of the overpressure region.
  8. 8. A 3D garment-based adaptive try-on display recommendation system, the system comprising: The initialization module is used for acquiring body data of a target user, constructing a three-dimensional human body model, and carrying out semantic region segmentation on the three-dimensional human body model to obtain a plurality of body semantic regions and corresponding local curvature characteristics respectively; The clothing instruction processing module is used for responding to the clothing selection instruction of the target user, retrieving a three-dimensional basic grid model of the target clothing, and calculating a target displacement vector corresponding to a preset characteristic control point on the three-dimensional basic grid model; The model parameter processing module is used for identifying grid vertexes corresponding to the body semantic regions in the three-dimensional basic grid model, determining deformation constraint weights of the grid vertexes according to the types of the body semantic regions, and calculating expansion coefficient adjustment values of the grid vertexes in the normal direction according to the local curvature characteristics; The model adjustment module is used for carrying out interpolation deformation calculation on the three-dimensional basic grid model according to the target displacement vector, the deformation constraint weight and the expansion coefficient adjustment value to generate a target clothing model adapting to the target user; the try-on display module is used for loading the target clothing model to the three-dimensional human body model and generating a dynamic try-on display picture according to the cloth physical attribute parameters corresponding to the target clothing; And the recommendation output module is used for generating a pressure distribution thermodynamic diagram according to the topological distance between the inner surface of the target clothing model and the outer surface of the three-dimensional human body model and generating clothing recommendation information based on the pressure distribution thermodynamic diagram and the dynamic try-in display picture.
  9. 9. A computer storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform the method of any one of claims 1 to 7.
  10. 10. An electronic device comprising a processor, a memory and a transceiver, the memory configured to store instructions, the transceiver configured to communicate with other devices, the processor configured to execute the instructions stored in the memory, to cause the electronic device to perform the method of any one of claims 1-7.

Description

Self-adaptive fitting display recommendation method and system based on 3D clothing Technical Field The application relates to the technical field of clothing data processing, in particular to a self-adaptive fitting display recommendation method and system based on 3D clothing. Background The virtual fitting technology is a technical means for simulating the wearing effect of clothing in a digital environment by utilizing computer graphics, three-dimensional modeling and image processing technologies. According to the technology, the human body three-dimensional model and the clothing model are constructed, so that virtual superposition and interactive display of clothing and human bodies are realized, and a user can preview the upper body effect of the clothing without actual try-on. The virtual try-on technology is widely applied to the fields of e-commerce platforms, clothing customization, fashion design and the like, provides convenient shopping decision support for users, reduces clothing return rate and improves shopping experience of the users. The common virtual try-on technology in the related art is mostly based on a two-dimensional patch mode, and the simulation try-on effect is achieved by covering the clothes picture on the human body model. However, in practical application, due to the reasons of the difference of body types, the complexity of body curves, the characteristics of clothing materials and the like, the fit of the clothing picture and the human body contour lacks real physical deformation, so that the user cannot accurately evaluate the comfort level and the fit degree of the clothing, and the try-on effect is poor. Disclosure of Invention The application provides a self-adaptive fitting display recommendation method and system based on 3D clothing, which can provide omnibearing real fitting experience for different users, thereby improving fitting effect. In a first aspect, the present application provides a 3D garment-based self-adaptive try-on display recommendation method, the method comprising: Acquiring body data of a target user, constructing a three-dimensional human body model, and carrying out semantic region segmentation on the three-dimensional human body model to obtain a plurality of body semantic regions and corresponding local curvature characteristics respectively; responding to a clothing selection instruction of a target user, retrieving a three-dimensional basic grid model of the target clothing, and calculating a target displacement vector corresponding to a preset characteristic control point on the three-dimensional basic grid model; Identifying grid vertexes corresponding to semantic areas of each body in the three-dimensional basic grid model, determining deformation constraint weights of the grid vertexes according to the types of the semantic areas of the body, and calculating expansion coefficient adjustment values of the grid vertexes in the normal direction according to local curvature characteristics; Performing interpolation deformation calculation on the three-dimensional basic grid model according to the target displacement vector, the deformation constraint weight and the expansion coefficient adjustment value to generate a target clothing model adapting to a target user; Loading the target clothing model into a three-dimensional human body model, and generating a dynamic try-on display picture according to the cloth physical attribute parameters corresponding to the target clothing; generating a pressure distribution thermodynamic diagram according to the topological distance between the inner surface of the target clothing model and the outer surface of the three-dimensional human body model, and generating clothing recommendation information based on the pressure distribution thermodynamic diagram and the dynamic try-on display picture. By adopting the technical scheme, the body data of the target user is obtained to construct a three-dimensional human body model, semantic region segmentation is carried out to obtain a plurality of body semantic regions and local curvature features, and personalized features of the body shape of the user and complexity of a body curve can be accurately captured. The deformation constraint weight of the grid vertexes is determined according to the type of the body semantic region, and the expansion coefficient adjustment value of the grid vertexes in the normal direction is calculated according to the local curvature characteristics, so that the garment model is accurately adapted to curvature changes of different body parts. And carrying out interpolation deformation calculation on the three-dimensional basic grid model through the target displacement vector, the deformation constraint weight and the expansion coefficient adjustment value to generate a target clothing model adapting to the target user, so that the clothing can truly fit with the body type characteristics of the user. The dynamic try-on displa