CoT3DRef: Chain-of-Thoughts Data-Efficient 3D Visual Grounding

arXiv:2310.06214v3 Announce Type: replace
Abstract: 3D visual grounding is the ability to localize objects in 3D scenes conditioned by utterances. Most existing methods devote the referring head to localize the referred object directly, causing failure in complex scenarios. In addition, it does not illustrate how and why the network reaches the final decision. In this paper, we address this question Can we design an interpretable 3D visual grounding framework that has the potential to mimic the human perception system?. To …

3d scenes abstract arxiv cs.cv data decision failure head network object objects thoughts type visual