Learning Geometric Representations from Videos for Spatial Intelligent Multimodal Large Language Models

Benchmark Model Rank Results
3d-question-answering-3d-qa-on-sqa3dGeoVR-4B#1Exact Match: 63.9
3d-question-answering-3d-qa-on-sqa3dGeoVR-2B#2Exact Match: 61.4
question-answering-on-sqa3dGeoVR-4B (ours)#1AnswerExactMatch (Question Answering): 63.9