Egocentric Video-Language Pretraining

Benchmark Model Rank Results
temporal-action-localization-on-ego4d-mq-validationVSGN + EgoVLP features#3mAP Avg IoU=0.1:0.5: 11.39R@1 IoU=0.5: 30.14
action-recognition-on-charades-egoEgoVLP#4mAP: 32.1
question-answering-on-egotaskqaEgoVLP#4Direct: 42.51
natural-language-queries-on-ego4dEgoVLP#9R@1 Mean(0.3 and 0.5): 8.35R@1 IoU=0.3: 10.46R@1 IoU=0.5: 6.24…