Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

YouCook

Benchmarks

Task NameDataset NameSOTA ResultTrend
Text-to-Video RetrievalYouCook2
Recall@1083.7
117
Video CaptioningYouCook II (val)
CIDEr221
98
Dense Video CaptioningYouCook2
SODA_c7.9
40
Video Level SummarizationYouCook2
METEOR30.08
21
Video RetrievalYoucook2
R@136.4
15
Action GroundingYouCook-Interactions (val)
Accuracy60.05
13
Temporal clip retrieval (video-to-text)YouCook
Recall@550
7
Temporal clip retrieval (text-to-video)YouCook
R@547.2
7
Dense Video CaptioningYouCook2 (test)
SODA_c1.2
6
Audio-to-video retrievalYouCook (test)
Recall@133.1
4
Object ClassificationYouCook II (val)
Object Top-1 Acc13.2
4
Verb ClassificationYouCook II (val)
Top-1 Accuracy0.161
4
Dense CaptioningYouCook2 zero-shot
SODA_c7.9
3
Video RetrievalYouCook II (test)
Avg Recall@{1,5,10}50.6
3
Video-Audio CaptioningYouCook2
CIDEr197.8
2
Object localizationYouCook2-BB
Full Localization0.5925
2
Video/Paragraph Retrieval (Video-to-Text)YouCook2
Recall@151.3
2
Showing 17 of 17 rows