Zero-Shot Learning on J-HMDB

Metric: Accuracy (RGB+pose) (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Accuracy (RGB+pose)▼	Augmentations	Paper	Date↕	Code
1	Potion	90.4	No	-	-	-
2	PA3D+RPAN	86.1	No	-	-	-
3	I3D + Potion	85.5	No	-	-	-
4	I3D	84.1	No	Quo Vadis, Action Recognition? A New Model and t...	2017-05-22	Code
5	RPAN	83.9	No	-	-	Code
6	Chained (RGB+Flow +Pose)	76.1	No	Chained Multi-stream Networks Exploiting Pose, M...	2017-04-03	Code
7	MR Two-Sream R-CNN	71.1	No	-	-	-
8	PA3D	69.5	No	-	-	-
9	STAR-Net	64.3	No	STAR-Net: Action Recognition using Spatio-Tempor...	2019-02-26	-
10	Action Tubes	62.5	No	Finding Action Tubes	2014-11-21	Code

#1Potion
90.4
Accuracy (RGB+pose)
No paper
#2PA3D+RPAN
86.1
Accuracy (RGB+pose)
No paper
#3I3D + Potion
85.5
Accuracy (RGB+pose)
No paper
#4I3DSOTA
84.1
Accuracy (RGB+pose)· 2017-05-22
Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset Code
#5RPAN
83.9
Accuracy (RGB+pose)
No paperCode
#6Chained (RGB+Flow +Pose)SOTA
76.1
Accuracy (RGB+pose)· 2017-04-03
Chained Multi-stream Networks Exploiting Pose, Motion, and Appearance for Action Classification and Detection Code
#7MR Two-Sream R-CNN
71.1
Accuracy (RGB+pose)
No paper
#8PA3D
69.5
Accuracy (RGB+pose)
No paper
#9STAR-Net
64.3
Accuracy (RGB+pose)· 2019-02-26
STAR-Net: Action Recognition using Spatio-Temporal Activation Reprojection
#10Action TubesSOTA
62.5
Accuracy (RGB+pose)· 2014-11-21
Finding Action Tubes Code