Method and device for function part division of three-dimensional model, electronic equipment and storage medium
By fusing point cloud data with multi-view 2D images through cross-modal attention and utilizing surface ID constraints, the problem of inaccurate differentiation of functional parts in 3D models was solved, achieving accurate identification of functional parts and improving CAE analysis.
CN121999337BActive Publication Date: 2026-06-26CHONGQING LANDIAN AUTOMOBILE TECHNOLOGY CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHONGQING LANDIAN AUTOMOBILE TECHNOLOGY CO LTD
- Filing Date
- 2026-04-07
- Publication Date
- 2026-06-26
AI Technical Summary
Technical Problem
Existing technologies cannot accurately distinguish the functional parts of a 3D model, resulting in a mismatch between the mesh generation and the actual structure of the parts, which affects the accuracy and efficiency of CAE analysis.
Method used
By acquiring point cloud data of a 3D model and multi-view 2D images, and using surface ID as the association benchmark, cross-modal attention fusion is performed. By combining point cloud detail features with the global structure of multi-view 2D images, accurate identification of functional parts can be achieved.
Benefits of technology
It enables accurate identification of each functional part of the 3D model, ensuring that the same surface is not incorrectly segmented, thus improving the accuracy and efficiency of CAE analysis.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN121999337B_ABST
Abstract
The application relates to a functional part division method and device of a three-dimensional model, electronic equipment and a storage medium. The method comprises the following steps: acquiring point cloud data and multi-view two-dimensional images of a three-dimensional model; based on the surface ID of each surface in the three-dimensional model, the surface ID of the surface to which each sampling point of the point cloud data belongs is bound, and the surface ID of the corresponding surface of the two-dimensional plane to which each pixel of the multi-view two-dimensional image belongs is bound; through a preset semantic segmentation model, the point cloud data and the multi-view two-dimensional images are cross-modal attention fused with the surface ID as the association reference, fused point cloud features are obtained, and the functional area label of each sampling point in the fused point cloud features is determined; the functional area label of each sampling point is mapped to the surface ID bound to the corresponding sampling point, and the functional area label corresponding to each surface ID is obtained; and the surface IDs with the same functional area label are merged into a surface set. The application realizes accurate identification of each functional part of the three-dimensional model.
Need to check novelty before this filing date? Find Prior Art