EgoVLPv2: Egocentric Video-Language Pre-training with Fusion in the Backbone
Open paper
Benchmark
Model
Rank
Results
action-recognition-on-charades-ego
EgoVLPv2
#2
mAP: 34.1
natural-language-queries-on-ego4d
EgoVLPv2
#10
R@1 IoU=0.3: 12.95
R@1 IoU=0.5: 7.91
R@5 IoU=0.3: 23.80
…
question-answering-on-egotaskqa
EgoVLPv2
#1
Direct: 46.26
Rank counts only results with a code link.