ViGiL3D: A Linguistically Diverse Dataset for 3D Visual Grounding

Publication
ArXiv