b
Discover
Models
Search
About
Towards CLIP-driven Language-free 3D Visual Grounding via 2D-3D Relational Enhancement and Consistency
5 months ago
·
CVPR