Zero-Shot Learning on THUMOS’14

Metric: mAP IOU@0.3 (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide augmentations

Sort:

#	Model↕	mAP IOU@0.3▼	Augmentations	Paper	Date↕	Code
1	AdaTAD (VideoMAEv2-giant)	89.7	No	End-to-End Temporal Action Detection with 1B Par...	2023-11-28	Code
2	RDFA-S6 (InternVideo2-6B)	88.7	No	Enhancing Temporal Action Localization: Advanced...	2024-07-18	Code
3	ActionMamba(InternVideo2-6B)	86.89	No	Video Mamba Suite: State Space Model as a Versat...	2024-03-14	Code
4	TriDet (VideoMAE v2-g feature)	84.8	Yes	Temporal Action Localization with Enhanced Insta...	2023-09-11	Code
5	ActionFormer (VideoMAE V2-g features)	84	Yes	VideoMAE V2: Scaling Video Masked Autoencoders w...	2023-03-29	Code
6	TriDet (I3D features)	83.6	No	TriDet: Temporal Action Detection with Relative ...	2023-03-13	Code
7	ASL(I3D features)	83.1	No	Action Sensitivity Learning for Temporal Action ...	2023-05-25	-
8	DualDETR (I3D features)	82.9	No	Dual DETRs for Multi-Label Temporal Action Detec...	2024-03-31	-
9	TemporalMaxer (I3D features)	82.8	No	TemporalMaxer: Maximize Temporal Context with on...	2023-03-16	Code
10	ActionFormer (I3D features)	82.1	No	ActionFormer: Localizing Moments of Actions with...	2022-02-16	Code
11	BasicTAD (160,6,192,R50-SlowOnly)	75.5	No	BasicTAD: an Astounding RGB-Only Baseline for Te...	2022-05-05	Code
12	TadTR	74.8	No	End-to-end Temporal Action Detection with Transf...	2021-06-18	Code
13	TadML(two-stream)	73.29	No	TadML: A fast temporal action detection with Mec...	2022-06-07	Code
14	AVFusion	70.1	Yes	Hear Me Out: Fusional Approaches for Audio Augme...	2021-06-27	Code
15	E2E-TAD (SlowFast R50+TadTR)	69.4	No	An Empirical Study of End-to-End Temporal Action...	2022-04-06	Code
16	ReAct (TSN features)	69.2	No	ReAct: Temporal Action Detection with Relational...	2022-07-14	Code
17	TSP	69.1	No	TSP: Temporally-Sensitive Pretraining of Video E...	2020-11-23	Code
18	MUSES	68.9	No	Multi-shot Temporal Event Localization: a Benchm...	2020-12-17	Code
19	TadML(rgb-only)	68.78	No	TadML: A fast temporal action detection with Mec...	2022-06-07	Code
20	TAGS (I3D)	68.6	No	Proposal-Free Temporal Action Detection via Glob...	2022-07-14	Code
21	BasicTAD (112,3,96,R50-SlowOnly)	68.4	No	BasicTAD: an Astounding RGB-Only Baseline for Te...	2022-05-05	Code
22	DCAN (TSN features)	68.2	No	DCAN: Improving Temporal Action Detection via Du...	2021-12-07	Code
23	VSGN	66.7	No	Video Self-Stitching Graph Network for Temporal ...	2020-11-30	Code
24	GCM	66.5	No	Graph Convolutional Module for Temporal Action L...	2021-12-01	-
25	AGT (Ours)	65	No	Activity Graph Transformer for Temporal Action L...	2021-01-21	-
26	P-GCN	63.6	No	Graph Convolutional Networks for Temporal Action...	2019-09-07	Code
27	DaoTAD	62.8	No	RGB Stream Is Enough for Temporal Action Detection	2021-07-09	Code
28	Decouple-SSAD	60.2	No	Decoupling Localization and Classification in Si...	2019-04-16	Code
29	ASM-Loc	57.1	No	ASM-Loc: Action-aware Segment Modeling for Weakl...	2022-03-29	Code
30	CO2-Net	54.5	No	Cross-modal Consensus Network for Weakly Supervi...	2021-07-27	Code
31	BSN UNet	53.5	No	BSN: Boundary Sensitive Network for Temporal Act...	2018-06-08	Code
32	TAL-Net	53.2	No	Rethinking the Faster R-CNN Architecture for Tem...	2018-04-20	-
33	CBR-TS	50.1	No	Cascaded Boundary Regression for Temporal Action...	2017-05-02	-
34	A2CL-PT	48.1	No	Adversarial Background-Aware Loss for Weakly-sup...	2020-07-13	Code
35	DeepMetricLearner	46.8	No	Weakly Supervised Temporal Action Localization U...	2020-01-21	Code
36	R-C3D	44.8	No	R-C3D: Region Convolutional 3D Network for Tempo...	2017-03-22	Code
37	TURN-FL-16 + S-CNN	44.1	No	TURN TAP: Temporal Unit Regression Network for T...	2017-03-17	Code
38	CDC	40.1	No	CDC: Convolutional-De-Convolutional Networks for...	2017-03-04	Code
39	S-CNN	36.3	No	Temporal Action Localization in Untrimmed Videos...	2016-01-09	Code
40	Yeung et al.	36	No	End-to-end Learning of Action Detection from Fra...	2015-11-22	Code

#1AdaTAD (VideoMAEv2-giant)SOTA
89.7
mAP IOU@0.3· 2023-11-28
End-to-End Temporal Action Detection with 1B Parameters Across 1000 Frames Code
#2RDFA-S6 (InternVideo2-6B)
88.7
mAP IOU@0.3· 2024-07-18
Enhancing Temporal Action Localization: Advanced S6 Modeling with Recurrent Mechanism Code
#3ActionMamba(InternVideo2-6B)
86.89
mAP IOU@0.3· 2024-03-14
Video Mamba Suite: State Space Model as a Versatile Alternative for Video Understanding Code
#4TriDet (VideoMAE v2-g feature)SOTA
84.8
mAP IOU@0.3· Augmentations· 2023-09-11
Temporal Action Localization with Enhanced Instant Discriminability Code
#5ActionFormer (VideoMAE V2-g features)SOTA
84
mAP IOU@0.3· Augmentations· 2023-03-29
VideoMAE V2: Scaling Video Masked Autoencoders with Dual Masking Code
#6TriDet (I3D features)SOTA
83.6
mAP IOU@0.3· 2023-03-13
TriDet: Temporal Action Detection with Relative Boundary Modeling Code
#7ASL(I3D features)
83.1
mAP IOU@0.3· 2023-05-25
Action Sensitivity Learning for Temporal Action Localization
#8DualDETR (I3D features)
82.9
mAP IOU@0.3· 2024-03-31
Dual DETRs for Multi-Label Temporal Action Detection
#9TemporalMaxer (I3D features)
82.8
mAP IOU@0.3· 2023-03-16
TemporalMaxer: Maximize Temporal Context with only Max Pooling for Temporal Action Localization Code
#10ActionFormer (I3D features)SOTA
82.1
mAP IOU@0.3· 2022-02-16
ActionFormer: Localizing Moments of Actions with Transformers Code
#11BasicTAD (160,6,192,R50-SlowOnly)
75.5
mAP IOU@0.3· 2022-05-05
BasicTAD: an Astounding RGB-Only Baseline for Temporal Action Detection Code
#12TadTRSOTA
74.8
mAP IOU@0.3· 2021-06-18
End-to-end Temporal Action Detection with Transformer Code
#13TadML(two-stream)
73.29
mAP IOU@0.3· 2022-06-07
TadML: A fast temporal action detection with Mechanics-MLP Code
#14AVFusion
70.1
mAP IOU@0.3· Augmentations· 2021-06-27
Hear Me Out: Fusional Approaches for Audio Augmented Temporal Action Localization Code
#15E2E-TAD (SlowFast R50+TadTR)
69.4
mAP IOU@0.3· 2022-04-06
An Empirical Study of End-to-End Temporal Action Detection Code
#16ReAct (TSN features)
69.2
mAP IOU@0.3· 2022-07-14
ReAct: Temporal Action Detection with Relational Queries Code
#17TSPSOTA
69.1
mAP IOU@0.3· 2020-11-23
TSP: Temporally-Sensitive Pretraining of Video Encoders for Localization Tasks Code
#18MUSES
68.9
mAP IOU@0.3· 2020-12-17
Multi-shot Temporal Event Localization: a Benchmark Code
#19TadML(rgb-only)
68.78
mAP IOU@0.3· 2022-06-07
TadML: A fast temporal action detection with Mechanics-MLP Code
#20TAGS (I3D)
68.6
mAP IOU@0.3· 2022-07-14
Proposal-Free Temporal Action Detection via Global Segmentation Mask Learning Code
#21BasicTAD (112,3,96,R50-SlowOnly)
68.4
mAP IOU@0.3· 2022-05-05
BasicTAD: an Astounding RGB-Only Baseline for Temporal Action Detection Code
#22DCAN (TSN features)
68.2
mAP IOU@0.3· 2021-12-07
DCAN: Improving Temporal Action Detection via Dual Context Aggregation Code
#23VSGN
66.7
mAP IOU@0.3· 2020-11-30
Video Self-Stitching Graph Network for Temporal Action Localization Code
#24GCM
66.5
mAP IOU@0.3· 2021-12-01
Graph Convolutional Module for Temporal Action Localization in Videos
#25AGT (Ours)
65
mAP IOU@0.3· 2021-01-21
Activity Graph Transformer for Temporal Action Localization
#26P-GCNSOTA
63.6
mAP IOU@0.3· 2019-09-07
Graph Convolutional Networks for Temporal Action Localization Code
#27DaoTAD
62.8
mAP IOU@0.3· 2021-07-09
RGB Stream Is Enough for Temporal Action Detection Code
#28Decouple-SSADSOTA
60.2
mAP IOU@0.3· 2019-04-16
Decoupling Localization and Classification in Single Shot Temporal Action Detection Code
#29ASM-Loc
57.1
mAP IOU@0.3· 2022-03-29
ASM-Loc: Action-aware Segment Modeling for Weakly-Supervised Temporal Action Localization Code
#30CO2-Net
54.5
mAP IOU@0.3· 2021-07-27
Cross-modal Consensus Network for Weakly Supervised Temporal Action Localization Code
#31BSN UNetSOTA
53.5
mAP IOU@0.3· 2018-06-08
BSN: Boundary Sensitive Network for Temporal Action Proposal Generation Code
#32TAL-NetSOTA
53.2
mAP IOU@0.3· 2018-04-20
Rethinking the Faster R-CNN Architecture for Temporal Action Localization
#33CBR-TSSOTA
50.1
mAP IOU@0.3· 2017-05-02
Cascaded Boundary Regression for Temporal Action Detection
#34A2CL-PT
48.1
mAP IOU@0.3· 2020-07-13
Adversarial Background-Aware Loss for Weakly-supervised Temporal Activity Localization Code
#35DeepMetricLearner
46.8
mAP IOU@0.3· 2020-01-21
Weakly Supervised Temporal Action Localization Using Deep Metric Learning Code
#36R-C3DSOTA
44.8
mAP IOU@0.3· 2017-03-22
R-C3D: Region Convolutional 3D Network for Temporal Activity Detection Code
#37TURN-FL-16 + S-CNNSOTA
44.1
mAP IOU@0.3· 2017-03-17
TURN TAP: Temporal Unit Regression Network for Temporal Action Proposals Code
#38CDCSOTA
40.1
mAP IOU@0.3· 2017-03-04
CDC: Convolutional-De-Convolutional Networks for Precise Temporal Action Localization in Untrimmed Videos Code
#39S-CNNSOTA
36.3
mAP IOU@0.3· 2016-01-09
Temporal Action Localization in Untrimmed Videos via Multi-stage CNNs Code
#40Yeung et al.SOTA
36
mAP IOU@0.3· 2015-11-22
End-to-end Learning of Action Detection from Frame Glimpses in Videos Code