Return
Boundary-aware shape recognition using dynamic graph convolutional networks
DOI:10.1016/j.patcog.2025.112511.png)
Abstract
En 中文
• Making the first effort on applying Graph Convolutional Networks (GCNs) to 2D shape recognition. • Proposing a Boundary-Aware Shape Recognition Graph Convolutional Network (BASR-GCN) which integrates both the boundary and skeleton cues. • Introducing a multi-scale BASR-GCN to extract shape features at different scales.
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W

