.png){kind=icon}
.png){kind=icon}
Return
PCNet: A composite backbone for 3D point cloud representation learning
.png){kind=icon}
PRE
AI
DOI：10.1016/j.patcog.2026.113521.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Proposing a composite framework that integrates 3D networks as a unified backbone. • Two fusion modules are developed to effectively aggregate multi-source features. • A 3D class activation mapping visualization for point localization is proposed.
Keywords:
3D point cloud
representation learning
fusion modules
class activation mapping
composite backbone
Journal
IF:
7.6
Papers:
1.3W
Citations:
4.5W
Organization
View more organizations