Skeleton Based Action Recognition
Skeleton-based Action Recognition is a computer vision task that focuses on recognizing and classifying human actions from sequences of 3D skeletal joint data captured by sensors. The goal of this task is to develop algorithms capable of understanding changes in human posture and accurately determining the type of action, with broad application prospects including human-computer interaction, motion analysis, and security monitoring.
CAD-120
Drive&Act
dyalyt
First-Person Hand Action Benchmark
TCN-Summ
Florence 3D
Gaming 3D (G3D)
H2O (2 Hands and Objects)
CHASE(STSA-Net)
HDM05
HMDB51
J-HMBD Early Action
DR^2N
J-HMDB
JHMDB (2D poses only)
HT-ConvNet
JHMDB Pose Tracking
mgPFF+ft 1st
Kinetics-400
STGAT
Kinetics-Skeleton dataset
PoseC3D (SlowOnly-346)
MSR Action3D
Temporal K-Means Clustering + Temporal Subspace Clustering
MSR ActionPairs
Temporal Subspace Clustering
MSRC-12
N-UCLA
SkateFormer
NTU RGB+D
PoseC3D [3D Heatmap]
NTU RGB+D 120
DeGCN
NTU60-X
4s-ShiftGCN
PKU-MMD
SBU / SBU-Refine
Joint Line Distance
SHREC 2017 track on 3D Hand Gesture Recognition
TD-GCN
Skeletics-152
4s-ShiftGCN
Skeleton-Mimetics
MS-G3D
SYSU 3D
SGN
TCG-dataset
Bidirectional LSTM
UAV-Human
HDBN
UCF101
UPenn Action
UNIK
UT-Kinect
Temporal Subspace Clustering
UWA3D
VA-fusion (aug.)
Varying-view RGB-D Action-Skeleton