3D-VisTA: Pre-trained Transformer for 3D Vision and Text Alignment
Open paper
Benchmark
Model
Rank
Results
3d-question-answering-3d-qa-on-scanqa-test-w
3D-VisTA
#7
Exact Match: 22.4
BLEU-4: 10.4
ROUGE: 35.7
METEOR: 13.9
CIDEr: 69.6
3d-question-answering-3d-qa-on-sqa3d
3D-VisTA
#5
Exact Match: 48.5
Rank counts only results with a code link.