基于局部和全局上下文感知的三维点云语义分割方法
By constructing a point cloud semantic segmentation network that fuses local and global features, the problem of insufficient utilization of local and global information in existing technologies is solved, achieving more efficient 3D point cloud semantic segmentation results and improving segmentation accuracy and category recognition capabilities.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HEBEI UNIV OF TECH
- Filing Date
- 2023-09-20
- Publication Date
- 2026-07-17
AI Technical Summary
Existing 3D point cloud semantic segmentation methods struggle to effectively utilize local and global contextual information when dealing with large-scale complex scenes, resulting in low segmentation accuracy and inaccurate semantic category assignment.
A point cloud semantic segmentation method based on local and global context awareness is adopted. Features are extracted through a local and global feature fusion module, and local and global features are learned by combining a local context encoding module and a dual attention mechanism. An end-to-end point cloud semantic segmentation network is constructed, and the KNN algorithm is used to obtain neighborhood point information for feature aggregation and weighted summation.
It improves the precision and accuracy of 3D point cloud semantic segmentation, reduces information loss, enhances the ability to analyze complex scenes, and achieves more efficient semantic category allocation.
Smart Images
Figure CN117218351B_ABST