Context-Aware Alignment and Mutual Masking for 3D-Language Pre-Training
Open paper
Benchmark
Model
Rank
Results
3d-dense-captioning-on-scanrefer-dataset
3D-VLP
#6
CIDEr: 50.02
BLEU-4: 31.87
METEOR: 24.53
ROUGE-L: 51.17
Rank counts only results with a code link.