GraFT: A Training-Free Framework for Spatial Reasoning in Multimodal Large Language Models via 3D Scene Graphs
GraFT is introduced, a training-free framework that supplies the missing 3D structure through a compact, easily maintained 3D scene graph (3DSG) and provides three spatial reasoning capabilities: deterministic geometry through symbolic tools, allocentric layout through a bird's-eye-view (BEV) rendering, and visual-attr...
Jun Du, Fernando Ropero, Erkin Turkoz et al.
· 0 citations