Towards CLIP-Driven Language-Free 3D Visual Grounding via 2D-3D Relational Enhancement and Consistency.

Yuqi Zhang, Han Luo, Yinjie Lei

Browse the full CVPR paper archive.