Video Understanding on Charades-STA

Metric: R5@0.7 (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	R5@0.7▼	Extra Data	Paper	Date↕	Code
1	DeCafNet	72.96	No	DeCafNet: Delegate and Conquer for Efficient Tem...	2025-05-22	Code
2	AdaFocus (Full, MViT-Charades-Pretrain-feature, MMN model)	66.4	No	Towards Weakly Supervised End-to-end Learning fo...	2023-11-28	-
3	AdaFocus (Full, I3D-Charades-Pretrain-feature, MMN model)	65	No	Towards Weakly Supervised End-to-end Learning fo...	2023-11-28	-
4	MMN (Full, MViT-K400-Pretrain-feature, evaluated by AdaFocus)	62.7	No	Negative Sample Matters: A Renaissance of Metric...	2021-09-10	Code
5	MMN (Full, I3D-K400-Pretrain-feature, evaluated by AdaFocus)	60.5	No	Negative Sample Matters: A Renaissance of Metric...	2021-09-10	Code
6	AdaFocus (Semi-weak, MViT-Charades-Pretrain-feature, D3G model)	54.6	No	Towards Weakly Supervised End-to-end Learning fo...	2023-11-28	-
7	AdaFocus (Weak, MViT-Charades-Pretrain-feature, CPL model)	52.6	No	Towards Weakly Supervised End-to-end Learning fo...	2023-11-28	-
8	AdaFocus (Weak, I3D-Charades-Pretrain-feature, CPL model)	51.8	No	Towards Weakly Supervised End-to-end Learning fo...	2023-11-28	-
9	CPL (Weak, MViT-K400-Pretrain-feature, evaluated by AdaFocus)	50.4	No	-	-	Code
10	D3G (Semi-weak, MViT-K400-Pretrain-feature, evaluated by AdaFocus)	50.2	No	D3G: Exploring Gaussian Prior for Temporal Sente...	2023-08-08	Code
11	AdaFocus (Semi-weak, I3D-Charades-Pretrain-feature, D3G model)	49.2	No	Towards Weakly Supervised End-to-end Learning fo...	2023-11-28	-
12	CPL (Weak, I3D-K400-Pretrain-feature, evaluated by AdaFocus)	49.2	No	-	-	Code
13	D3G (Semi-weak, I3D-K400-Pretrain-feature, evaluated by AdaFocus)	48	No	D3G: Exploring Gaussian Prior for Temporal Sente...	2023-08-08	Code

#1DeCafNetSOTA
72.96
R5@0.7· 2025-05-22
DeCafNet: Delegate and Conquer for Efficient Temporal Grounding in Long Videos Code
#2AdaFocus (Full, MViT-Charades-Pretrain-feature, MMN model)SOTA
66.4
R5@0.7· 2023-11-28
Towards Weakly Supervised End-to-end Learning for Long-video Action Recognition
#3AdaFocus (Full, I3D-Charades-Pretrain-feature, MMN model)
65
R5@0.7· 2023-11-28
Towards Weakly Supervised End-to-end Learning for Long-video Action Recognition
#4MMN (Full, MViT-K400-Pretrain-feature, evaluated by AdaFocus)SOTA
62.7
R5@0.7· 2021-09-10
Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding Code
#5MMN (Full, I3D-K400-Pretrain-feature, evaluated by AdaFocus)
60.5
R5@0.7· 2021-09-10
Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding Code
#6AdaFocus (Semi-weak, MViT-Charades-Pretrain-feature, D3G model)
54.6
R5@0.7· 2023-11-28
Towards Weakly Supervised End-to-end Learning for Long-video Action Recognition
#7AdaFocus (Weak, MViT-Charades-Pretrain-feature, CPL model)
52.6
R5@0.7· 2023-11-28
Towards Weakly Supervised End-to-end Learning for Long-video Action Recognition
#8AdaFocus (Weak, I3D-Charades-Pretrain-feature, CPL model)
51.8
R5@0.7· 2023-11-28
Towards Weakly Supervised End-to-end Learning for Long-video Action Recognition
#9CPL (Weak, MViT-K400-Pretrain-feature, evaluated by AdaFocus)
50.4
R5@0.7
No paperCode
#10D3G (Semi-weak, MViT-K400-Pretrain-feature, evaluated by AdaFocus)
50.2
R5@0.7· 2023-08-08
D3G: Exploring Gaussian Prior for Temporal Sentence Grounding with Glance Annotation Code
#11AdaFocus (Semi-weak, I3D-Charades-Pretrain-feature, D3G model)
49.2
R5@0.7· 2023-11-28
Towards Weakly Supervised End-to-end Learning for Long-video Action Recognition
#12CPL (Weak, I3D-K400-Pretrain-feature, evaluated by AdaFocus)
49.2
R5@0.7
No paperCode
#13D3G (Semi-weak, I3D-K400-Pretrain-feature, evaluated by AdaFocus)
48
R5@0.7· 2023-08-08
D3G: Exploring Gaussian Prior for Temporal Sentence Grounding with Glance Annotation Code