SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities

Internet-scale 3D spatial reasoning dataset in metric space. By training a VLM on such data, we significantly enhance its ability on both qualitative and quantitative spatial VQA. Finally, we demonstrate that this VLM unlocks novel downstream applications in chainof-thought spatial reasoning and robotics due to its quantitative estimation capability. Website: https://spatial-vlm.github.io/