Skeleton Based Action Recognition On Kinetics

Metrics

Accuracy

Results

Performance results of various models on this benchmark

Model Name	Accuracy	Paper Title	Repository
ProtoGCN	51.9	Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition
ST-TR-agcn	37.4	Skeleton-based Action Recognition via Spatial and Temporal Transformer Networks
ST-GCN (2-stream)	34.4	Continual Spatio-Temporal Graph Convolutional Networks
CGCN	37.5	Unifying Graph Embedding Features with Graph Convolutional Networks for Skeleton-based Action Recognition	-
CoST-GCN* (1-stream)	30.2	Continual Spatio-Temporal Graph Convolutional Networks
Structured Keypoint Pooling (PPNv2 skeletons+objects)	52.3	Unified Keypoint-based Action Recognition Framework via Structured Keypoint Pooling	-
CoAGCN* (1-stream)	23.3	Continual Spatio-Temporal Graph Convolutional Networks
SLnL-rFA	36.6	Skeleton-Based Action Recognition with Synchronous Local and Non-local Spatio-temporal Learning and Frequency Attention	-
ST-GR	33.6	Spatiotemporal graph routing for skeleton-based action recognition	-
DualHead-Net	38.4	Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based Action Recognition
Dynamic GCN	37.9	Dynamic GCN: Context-enriched Topology Learning for Skeleton-based Action Recognition
CoST-GCN* (2-stream)	32.2	Continual Spatio-Temporal Graph Convolutional Networks
MS-G3D	38.0	Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition
JB-AAGCN	37.4	Skeleton-Based Action Recognition with Multi-Stream Adaptive Graph Convolutional Networks
AS-GCN	34.8	Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognition
DGNN	36.9	Skeleton-Based Action Recognition With Directed Graph Neural Networks	-
CoS-TR* (1-stream)	27.4	Continual Spatio-Temporal Graph Convolutional Networks
CoAGCN (1-stream)	33	Continual Spatio-Temporal Graph Convolutional Networks
2s-AGCN+TEM	38.6	Temporal Extension Module for Skeleton-Based Action Recognition	-
HD-GCN	40.9	Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition

0 of 42 row(s) selected.