GPT4Scene: Understand 3D Scenes from Videos with Vision-Language Models
Open paper
Benchmark
Model
Rank
Results
3d-question-answering-3d-qa-on-scanqa-val
Qwen2-VL-7B (GPT4Scene)-HDM
#1
CIDEr: 96.3
3d-visual-grounding-on-scanrefer-val
Qwen2-VL-7B (GPT4Scene)-HDM
#3
Acc@0.25: 62.6
Rank counts only results with a code link.