Home Console Docs News Papers Tutorials Datasets Wiki SOTA LLM Models GPU Leaderboard Events

English

Action Detection On Ucf101 24

Metrics

Video-mAP 0.2

Video-mAP 0.5

Results

Performance results of various models on this benchmark

			Paper Title	Repository
Stable Mean Teacher (I3D)	-	76.3	Stable Mean Teacher for Semi-supervised Video Action Detection
HIT	88.8	74.3	Holistic Interaction Transformer Network for Action Detection
E2E-SSL (I3D)	-	72.1	End-to-End Semi-Supervised Learning for Video Action Detection
STAR/L	88.0	71.8	End-to-End Spatio-Temporal Action Localisation with Video Transformers	-
Faster-RCNN + two-stream I3D conv	-	59.9	AVA: A Video Dataset of Spatio-temporally Localized Atomic Visual Actions
DTS	-	54	Finding Action Tubes with a Sparse-to-Dense Framework	-
MOC	81.8	53.9	Actions as Moving Points
YOWO + LFB	78.6	53.1	You Only Watch Once: A Unified CNN Architecture for Real-Time Spatiotemporal Action Localization
TACNet	77.5	52.9	TACNet: Transition-Aware Context Network for Spatio-Temporal Action Detection	-
HISAN (ResNet-101 + FPN)	82.30	51.47	Hierarchical Self-Attention Network for Action Localization in Videos	-
Two-in-one Two Stream	78.48	50.30	Dance with Flow: Two-in-One Stream Action Detection
HISAN (VGG-16)	80.42	49.50	Hierarchical Self-Attention Network for Action Localization in Videos	-
YOWO	75.8	48.8	You Only Watch Once: A Unified CNN Architecture for Real-Time Spatiotemporal Action Localization
Two-in-one	75.48	48.31	Dance with Flow: Two-in-One Stream Action Detection
MR-TS R-CNN	-	-	Multi-region two-stream R-CNN for action detection	-
TS R-CNN	-	-	Multi-region two-stream R-CNN for action detection	-
STEP	76.6	-	STEP: Spatio-Temporal Progressive Learning for Video Action Detection
T-CNN	47.1	-	Tube Convolutional Neural Network (T-CNN) for Action Detection in Videos

0 of 18 row(s) selected.