Activity Recognition on NTU RGB+D

Metric: Accuracy (CS) (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	Accuracy (CS)▼	Extra Data	Paper	Date↕	Code
1	DSCNet (RGB + Pose)	97.4	No	-	-	Code
2	PoseC3D (RGB + Pose)	97	No	Revisiting Skeleton-based Action Recognition	2021-04-28	Code
3	π-ViT (RGB + Pose)	96.3	Yes	Just Add $π$! Pose Induced Video Transformers fo...	2023-11-30	Code
4	UMDR (RGB-D)	96.2	No	A Unified Multimodal De- and Re-coupling Framewo...	2022-11-16	Code
5	EPAM-Net	96.1	No	EPAM-Net: An Efficient Pose-driven Attention-gui...	2024-08-10	Code
6	MMNet (RGB + Pose)	96	Yes	-	-	Code
7	Hierarchical Action Classification (RGB + Pose)	95.66	No	Hierarchical Action Classification with Network ...	2020-07-30	-
8	VPN (RGB + Pose)	95.5	Yes	VPN: Learning Video-Pose Embedding for Activitie...	2020-07-06	Code
9	EPP-Net (Parsing + Pose)	94.7	No	Explore Human Parsing Modality for Action Recogn...	2024-01-04	Code
10	3DA (RGB + Pose)	94.3	No	Cross-Modal Learning with 3D Deformable Attentio...	2022-12-12	-
11	Action Machine (RGB only)	94.3	No	Action Machine: Rethinking Action Recognition in...	2018-12-14	-
12	Hulk(Finetune, ViT-L)	94.3	No	Hulk: A Universal Knowledge Translator for Human...	2023-12-04	Code
13	PoseC3D [3D Heatmap]	94.1	Yes	Revisiting Skeleton-based Action Recognition	2021-04-28	Code
14	π-ViT (RGB only)	94	Yes	Just Add $π$! Pose Induced Video Transformers fo...	2023-11-30	Code
15	Hulk(Finetune, ViT-B)	94	No	Hulk: A Universal Knowledge Translator for Human...	2023-12-04	Code
16	MaskCLR	93.9	No	-	-	-
17	IPP-Net (Parsing + Pose)	93.8	No	Integrating Human Parsing and Pose Network for H...	2023-07-16	Code
18	ProtoGCN	93.8	No	Revealing Key Details to See Differences: A Nove...	2024-11-28	Code
19	ViewCon (RGB + Pose)	93.7	No	-	-	Code
20	JMDA (based on Skeleton MixFormer)	93.7	No	-	-	Code
21	Shap-Mix	93.7	No	Shap-Mix: Shapley Value Guided Mixing for Long-T...	2024-07-17	Code
22	MSA-GCN	93.6	No	-	-	-
23	DeGCN	93.6	No	-	-	Code
24	FreqMixFormer	93.6	No	Frequency Guidance Matters: Skeletal Action Reco...	2024-07-17	Code
25	SkateFormer	93.5	No	SkateFormer: Skeletal-Temporal Transformer for H...	2024-03-14	Code
26	MMCL	93.5	Yes	Multi-Modality Co-Learning for Efficient Skeleto...	2024-07-22	Code
27	LA-GCN	93.5	Yes	Language Knowledge-Assisted Representation Learn...	2023-05-21	Code
28	DVANet (RGB only)	93.4	No	DVANet: Disentangling View and Action Features f...	2023-12-10	Code
29	HD-GCN	93.4	No	Hierarchically Decomposed Graph Convolutional Ne...	2022-08-23	Code
30	JPFormer	93.2	No	-	-	Code
31	DG-STGCN	93.2	No	DG-STGCN: Dynamic Spatial-Temporal Modeling for ...	2022-10-12	Code
32	STEP-CATFormer	93.2	Yes	STEP CATFormer: Spatial-Temporal Effective Body-...	2023-12-06	Code
33	JPFormer	93.2	No	-	-	Code
34	MAMP	93.1	No	Masked Motion Predictors are Strong 3D Action Re...	2023-08-14	Code
35	SkeletonGCL (based on CTR-GCN)	93.1	No	Graph Contrastive Learning for Skeleton-based Ac...	2023-01-26	Code
36	BlockGCN	93.1	No	-	-	Code
37	MotionBert (finetune)	93	No	MotionBERT: A Unified Perspective on Learning Hu...	2022-10-12	Code
38	InfoGCN	93	No	-	-	Code
39	LST	92.9	Yes	Generative Action Description Prompts for Skelet...	2022-08-10	Code
40	Stream-GCN	92.9	No	Action Recognition with Multi-stream Motion Mode...	2023-06-13	-
41	PSUMNet	92.9	No	PSUMNet: Unified Modality Part Streams are All Y...	2022-08-11	Code
42	Hyperformer	92.9	No	Hypergraph Transformer for Skeleton-based Action...	2022-11-17	Code
43	STGAT	92.8	No	Spatial Temporal Graph Attention Network for Ske...	2022-08-18	Code
44	TCA-GCN	92.8	No	Skeleton-based Action Recognition via Temporal-C...	2022-05-31	Code
45	TD-GCN	92.8	No	-	-	Code
46	DSTSA-GCN	92.78	No	DSTSA-GCN: Advancing Skeleton-Based Gesture Reco...	2025-01-21	Code
47	DSTSA-GCN	92.78	No	DSTSA-GCN: Advancing Skeleton-Based Gesture Reco...	2025-01-21	Code
48	MSSTNet	92.6	No	-	-	Code
49	ST-GCN++ [PYSKL, 3D Skeleton]	92.6	No	PYSKL: Towards Good Practices for Skeleton Actio...	2022-05-19	Code
50	TSMF (RGB + Pose)	92.5	Yes	-	-	Code
51	CTR-GCN	92.4	No	Channel-wise Topology Refinement Graph Convoluti...	2021-07-26	Code
52	MSAF (RGB+Pose)	92.24	No	MSAF: Multimodal Split Attention Fusion	2020-12-13	Code
53	EfficientGCN-B4	92.1	No	Constructing Stronger and Faster Baselines for S...	2021-06-29	Code
54	STAR-Transformer (RGB + Pose)	92	No	STAR-Transformer: A Spatio-temporal Cross Attent...	2022-10-14	-
55	DualHead-Net	92	No	Learning Multi-Granular Spatio-Temporal Graph Ne...	2021-08-10	Code
56	MMTM (RGB+Pose)	91.99	No	MMTM: Multimodal Transfer Module for CNN Fusion	2019-11-20	Code
57	FUSION (IR+Pose)	91.8	No	Infrared and 3D skeleton feature fusion for RGB-...	2020-02-28	Code
58	PoseMap (RGB+Pose)	91.7	No	-	-	-
59	B2C-AFM(RGB+Pose)	91.7	No	-	-	Code
60	AngNet-JA + BA + JBA + VJBA	91.7	Yes	Fusing Higher-order Features in Graph Neural Net...	2021-05-04	Code
61	Skeletal GNN	91.6	Yes	Learning Skeletal Graph Neural Networks for Hard...	2021-08-16	-
62	SCK⊕	91.56	Yes	Tensor Representations for Action Recognition	2020-12-28	Code
63	DSTA-Net	91.5	No	Decoupled Spatial-Temporal Attention Network for...	2020-07-07	Code
64	MS-G3D Net	91.5	Yes	Disentangling and Unifying Graph Convolutions fo...	2020-03-31	Code
65	Dynamic GCN	91.5	No	Dynamic GCN: Context-enriched Topology Learning ...	2020-07-29	Code
66	ST-GCN [PYSKL, 2D Skeleton]	91.4	No	PYSKL: Towards Good Practices for Skeleton Actio...	2022-05-19	Code
67	MS-AAGCN+TEM	91	No	Temporal Extension Module for Skeleton-Based Act...	2020-03-19	-
68	PA-ResGCN-B19	90.9	No	Stronger, Faster and More Explainable: A Graph C...	2020-10-20	Code
69	EfficientGCN-B2	90.9	No	Constructing Stronger and Faster Baselines for S...	2021-06-29	Code
70	4s Shift-GCN	90.7	Yes	-	-	Code
71	ST-GCN [PYSKL, 3D Skeleton]	90.7	No	Spatial Temporal Graph Convolutional Networks fo...	2018-01-23	Code
72	CGCN	90.3	No	Unifying Graph Embedding Features with Graph Con...	2020-03-06	-
73	BAGCN	90.3	No	Focusing and Diffusion: Bidirectional Attentive ...	2019-12-24	-
74	FGCN-spatial+FGCN-motion	90.2	No	Feedback Graph Convolutional Network for Skeleto...	2020-03-17	-
75	Sym-GNN	90.1	No	Symbiotic Graph Neural Networks for 3D Skeleton-...	2019-10-05	-
76	ST-GCN [Vanilla, 2D Skeleton]	90.1	No	Spatial Temporal Graph Convolutional Networks fo...	2018-01-23	Code
77	Action Capsules	90	No	Action Capsules: Human Skeleton Action Recognition	2023-01-30	-
78	MS-AAGCN	90	No	Skeleton-Based Action Recognition with Multi-Str...	2019-12-15	Code
79	DGNN	89.9	No	-	-	Code
80	ST-TR-agcn	89.9	No	Skeleton-based Action Recognition via Spatial an...	2020-08-17	Code
81	EfficientGCN-B0	89.9	No	Constructing Stronger and Faster Baselines for S...	2021-06-29	Code
82	Mix-Dimension	89.7	No	Mix Dimension in Poincaré Geometry for 3D Skelet...	2020-07-30	-
83	2s-SDGCN	89.58	No	-	-	-
84	TD-GDSCN	89.57	No	-	-	Code
85	JB-AAGCN	89.4	No	Skeleton-Based Action Recognition with Multi-Str...	2019-12-15	Code
86	GCN-NAS	89.4	No	Learning Graph Convolutional Network for Skeleto...	2019-11-11	Code
87	VA-fusion (aug.)	89.4	No	View Adaptive Neural Networks for High Performan...	2018-04-20	Code
88	AGC-LSTM (Joint&Part)	89.2	No	An Attention Enhanced Graph Convolutional LSTM N...	2019-02-25	-
89	3s-HYSP	89.1	No	HYperbolic Self-Paced Learning for Self-Supervis...	2023-03-10	Code
90	SLnL-rFA	89.1	No	Skeleton-Based Action Recognition with Synchrono...	2018-11-10	-
91	SGN	89	No	Semantics-Guided Neural Networks for Efficient S...	2019-04-02	Code
92	CoS-TR* (2-stream)	88.9	No	Continual Spatio-Temporal Graph Convolutional Ne...	2022-03-21	Code
93	3SCNN	88.6	No	-	-	-
94	2s-NLGCN	88.5	No	Two-Stream Adaptive Graph Convolutional Networks...	2018-05-20	Code
95	AutoGCN	88.3	No	AutoGCN -- Towards Generic Human Activity Recogn...	2024-02-02	Code
96	CoST-GCN* (2-stream)	88.3	No	Continual Spatio-Temporal Graph Convolutional Ne...	2022-03-21	Code
97	PGCN-TCA	88	No	-	-	-
98	PB-GCN (Skeleton only)	87.5	No	Part-based Graph Convolutional Network for Actio...	2018-09-13	Code
99	GR-GCN	87.5	No	Optimized Skeleton-based Action Recognition via ...	2018-11-29	-
100	PB-GCN	87.5	No	Part-based Graph Convolutional Network for Actio...	2018-09-13	Code
101	3s RA-GCN	87.3	No	Richly Activated Graph Convolutional Network for...	2020-08-09	Code
102	TS-SAN	87.2	No	Self-Attention Network for Skeleton-based Human ...	2019-12-18	-
103	3s-USDRL (DSTE) This work	87.1	No	USDRL: Unified Skeleton-Based Dense Representati...	2024-12-12	Code
104	AS-GCN	86.8	No	Actional-Structural Graph Convolutional Networks...	2019-04-26	Code
105	RF-Action	86.8	No	Making the Invisible Visible: Action Recognition...	2019-09-20	-
106	Dense IndRNN	86.7	No	Deep Independently Recurrent Neural Network (Ind...	2019-10-11	Code
107	Glimpse Clouds (RGB only)	86.6	No	Glimpse Clouds: Human Activity Recognition from ...	2018-02-22	Code
108	ST-GCN [Vanilla, 3D Skeleton]	86.6	No	Spatial Temporal Graph Convolutional Networks fo...	2018-01-23	Code
109	HYSP	86.5	No	HYperbolic Self-Paced Learning for Self-Supervis...	2023-03-10	Code
110	HCN	86.5	No	Co-occurrence Feature Learning from Skeleton Dat...	2018-04-17	Code
111	CoST-GCN*	86.3	No	Continual Spatio-Temporal Graph Convolutional Ne...	2022-03-21	Code
112	CoS-TR*	86.3	No	Continual Spatio-Temporal Graph Convolutional Ne...	2022-03-21	Code
113	Sem-GCN	86.2	No	-	-	-
114	ST-GCN	86	No	Continual Spatio-Temporal Graph Convolutional Ne...	2022-03-21	Code
115	CoAGCN* (2-stream)	86	No	Continual Spatio-Temporal Graph Convolutional Ne...	2022-03-21	Code
116	3s RA-GCN	85.9	No	Richly Activated Graph Convolutional Network for...	2019-05-16	Code
117	2s RA-GCN	85.8	No	Richly Activated Graph Convolutional Network for...	2019-05-16	Code
118	PeGCN	85.6	No	Predictively Encoded Graph Convolutional Network...	2020-03-17	Code
119	2D-3D-Softargmax (RGB only)	85.5	No	2D/3D Pose Estimation and Action Recognition usi...	2018-02-26	Code
120	GVFE + AS-GCN with DH-TCN	85.3	No	Vertex Feature Encoding and Hierarchical Tempora...	2019-12-20	-
121	PR-GCN	85.2	No	Pose Refinement Graph Convolutional Network for ...	2020-10-14	-
122	AR-GCN	85.1	No	-	-	-
123	3scale ResNet152	85	No	Skeleton based action recognition using translat...	2017-04-19	-
124	SR-TSL	84.8	No	Skeleton-Based Action Recognition with Spatial R...	2018-05-07	-
125	3s-ActCLR	84.3	No	Actionlet-Dependent Contrastive Learning for Uns...	2023-03-20	-
126	RGB+Skeleton (cross-attention)	84.23	No	-	-	-
127	CoAGCN*	84.1	No	Continual Spatio-Temporal Graph Convolutional Ne...	2022-03-21	Code
128	ST-GCN-jpd	83.36	No	A Comparative Review of Recent Kinect-based Acti...	2019-06-24	Code
129	CNN+Motion+Trans	83.2	No	Skeleton-based Action Recognition with Convoluti...	2017-04-25	Code
130	IndRNN (with jpd)	83	No	A Comparative Review of Recent Kinect-based Acti...	2019-06-24	Code
131	FO-GASTM	82.83	No	-	-	-
132	MANs (DenseNet-161)	82.67	No	Memory Attention Networks for Skeleton-based Act...	2018-04-23	Code
133	DM-3DCNN	82	No	-	-	Code
134	Bayesian GC-LSTM	81.8	No	-	-	-
135	Ind-RNN	81.8	No	Independently Recurrent Neural Network (IndRNN):...	2018-03-13	Code
136	ST-GCN	81.5	No	Spatial Temporal Graph Convolutional Networks fo...	2018-01-23	Code
137	HPM_RGB+HPM_3D+Traj	80.9	No	Learning Human Pose Models from Synthesized Data...	2017-07-04	-
138	ARRN-LSTM	80.7	No	Relational Network for Skeleton-Based Action Rec...	2018-05-07	-
139	EleAtt-GRU (aug.)	80.7	No	EleAtt-RNN: Adding Attentiveness to Neurons in R...	2019-09-03	-
140	Skeleton Visualization	80	No	-	-	-
141	EleAtt-GRU	79.8	No	Adding Attentiveness to the Neurons in Recurrent...	2018-07-12	-
142	Clips+CNN+MTLN	79.6	No	A New Representation of Skeleton Sequences for 3...	2017-03-09	-
143	F2CSkeleton	79.6	No	A Fine-to-Coarse Convolutional Neural Network fo...	2018-05-30	-
144	VA-LSTM	79.2	No	View Adaptive Recurrent Neural Networks for High...	2017-03-24	Code
145	Skelemotion + Yang et al. (Skeleton only)	76.5	No	SkeleMotion: A New Representation of Skeleton Jo...	2019-07-30	Code
146	Skelemotion + Yang et al.	76.5	No	SkeleMotion: A New Representation of Skeleton Jo...	2019-07-30	Code
147	GCA-LSTM	76.1	No	-	-	-
148	SkeletonNet	75.9	No	-	-	-
149	DSSCA-SSLM (RGB only)	74.9	No	Deep Multimodal Feature Analysis for Action Reco...	2016-03-23	-
150	URNN-2L-T	74.6	No	-	-	-
151	Ensemble TS-LSTM v2	74.6	No	-	-	Code
152	TCN	74.3	No	Interpretable 3D Human Action Analysis with Temp...	2017-04-14	Code
153	STA-LSTM	73.4	No	An End-to-End Spatio-Temporal Attention Model fo...	2016-11-18	-
154	TSRJI (Late Fusion) + HCN	73.3	No	Skeleton Image Representation for 3D Action Reco...	2019-09-11	Code
155	Two-Stream RNN	71.3	No	Modeling Temporal Dynamics and Spatial Configura...	2017-04-09	-
156	Spatio-Temporal LSTM	69.2	No	Spatio-Temporal LSTM with Trust Gates for 3D Hum...	2016-07-24	-
157	Part-aware LSTM	62.93	No	NTU RGB+D: A Large Scale Dataset for 3D Human Ac...	2016-04-11	Code
158	ST-LSTM	61.7	No	Spatio-Temporal LSTM with Trust Gates for 3D Hum...	2016-07-24	-
159	Deep LSTM	60.7	No	NTU RGB+D: A Large Scale Dataset for 3D Human Ac...	2016-04-11	Code
160	Dynamic Skeletons	60.2	No	-	-	-
161	H-RNN	59.1	No	-	-	-
162	Lie Group	50.1	No	-	-	Code
163	Skeleton Quads	38.6	No	-	-	-

#1DSCNet (RGB + Pose)
97.4
Accuracy (CS)
No paperCode
#2PoseC3D (RGB + Pose)SOTA
97
Accuracy (CS)· 2021-04-28
Revisiting Skeleton-based Action Recognition Code
#3π-ViT (RGB + Pose)
96.3
Accuracy (CS)· Extra Data· 2023-11-30
Just Add $π$! Pose Induced Video Transformers for Understanding Activities of Daily Living Code
#4UMDR (RGB-D)
96.2
Accuracy (CS)· 2022-11-16
A Unified Multimodal De- and Re-coupling Framework for RGB-D Motion Recognition Code
#5EPAM-Net
96.1
Accuracy (CS)· 2024-08-10
EPAM-Net: An Efficient Pose-driven Attention-guided Multimodal Network for Video Action Recognition Code
#6MMNet (RGB + Pose)
96
Accuracy (CS)· Extra Data
No paperCode
#7Hierarchical Action Classification (RGB + Pose)SOTA
95.66
Accuracy (CS)· 2020-07-30
Hierarchical Action Classification with Network Pruning
#8VPN (RGB + Pose)SOTA
95.5
Accuracy (CS)· Extra Data· 2020-07-06
VPN: Learning Video-Pose Embedding for Activities of Daily Living Code
#9EPP-Net (Parsing + Pose)
94.7
Accuracy (CS)· 2024-01-04
Explore Human Parsing Modality for Action Recognition Code
#103DA (RGB + Pose)
94.3
Accuracy (CS)· 2022-12-12
Cross-Modal Learning with 3D Deformable Attention for Action Recognition
#11Action Machine (RGB only)SOTA
94.3
Accuracy (CS)· 2018-12-14
Action Machine: Rethinking Action Recognition in Trimmed Videos
#12Hulk(Finetune, ViT-L)
94.3
Accuracy (CS)· 2023-12-04
Hulk: A Universal Knowledge Translator for Human-Centric Tasks Code
#13PoseC3D [3D Heatmap]
94.1
Accuracy (CS)· Extra Data· 2021-04-28
Revisiting Skeleton-based Action Recognition Code
#14π-ViT (RGB only)
94
Accuracy (CS)· Extra Data· 2023-11-30
Just Add $π$! Pose Induced Video Transformers for Understanding Activities of Daily Living Code
#15Hulk(Finetune, ViT-B)
94
Accuracy (CS)· 2023-12-04
Hulk: A Universal Knowledge Translator for Human-Centric Tasks Code
#16MaskCLR
93.9
Accuracy (CS)
No paper
#17IPP-Net (Parsing + Pose)
93.8
Accuracy (CS)· 2023-07-16
Integrating Human Parsing and Pose Network for Human Action Recognition Code
#18ProtoGCN
93.8
Accuracy (CS)· 2024-11-28
Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition Code
#19ViewCon (RGB + Pose)
93.7
Accuracy (CS)
No paperCode
#20JMDA (based on Skeleton MixFormer)
93.7
Accuracy (CS)
No paperCode
#21Shap-Mix
93.7
Accuracy (CS)· 2024-07-17
Shap-Mix: Shapley Value Guided Mixing for Long-Tailed Skeleton Based Action Recognition Code
#22MSA-GCN
93.6
Accuracy (CS)
No paper
#23DeGCN
93.6
Accuracy (CS)
No paperCode
#24FreqMixFormer
93.6
Accuracy (CS)· 2024-07-17
Frequency Guidance Matters: Skeletal Action Recognition by Frequency-Aware Mixed Transformer Code
#25SkateFormer
93.5
Accuracy (CS)· 2024-03-14
SkateFormer: Skeletal-Temporal Transformer for Human Action Recognition Code
#26MMCL
93.5
Accuracy (CS)· Extra Data· 2024-07-22
Multi-Modality Co-Learning for Efficient Skeleton-based Action Recognition Code
#27LA-GCN
93.5
Accuracy (CS)· Extra Data· 2023-05-21
Language Knowledge-Assisted Representation Learning for Skeleton-Based Action Recognition Code
#28DVANet (RGB only)
93.4
Accuracy (CS)· 2023-12-10
DVANet: Disentangling View and Action Features for Multi-View Action Recognition Code
#29HD-GCN
93.4
Accuracy (CS)· 2022-08-23
Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition Code
#30JPFormer
93.2
Accuracy (CS)
No paperCode
#31DG-STGCN
93.2
Accuracy (CS)· 2022-10-12
DG-STGCN: Dynamic Spatial-Temporal Modeling for Skeleton-based Action Recognition Code
#32STEP-CATFormer
93.2
Accuracy (CS)· Extra Data· 2023-12-06
STEP CATFormer: Spatial-Temporal Effective Body-Part Cross Attention Transformer for Skeleton-based Action Recognition Code
#33JPFormer
93.2
Accuracy (CS)
No paperCode
#34MAMP
93.1
Accuracy (CS)· 2023-08-14
Masked Motion Predictors are Strong 3D Action Representation Learners Code
#35SkeletonGCL (based on CTR-GCN)
93.1
Accuracy (CS)· 2023-01-26
Graph Contrastive Learning for Skeleton-based Action Recognition Code
#36BlockGCN
93.1
Accuracy (CS)
No paperCode
#37MotionBert (finetune)
93
Accuracy (CS)· 2022-10-12
MotionBERT: A Unified Perspective on Learning Human Motion Representations Code
#38InfoGCN
93
Accuracy (CS)
No paperCode
#39LST
92.9
Accuracy (CS)· Extra Data· 2022-08-10
Generative Action Description Prompts for Skeleton-based Action Recognition Code
#40Stream-GCN
92.9
Accuracy (CS)· 2023-06-13
Action Recognition with Multi-stream Motion Modeling and Mutual Information Maximization
#41PSUMNet
92.9
Accuracy (CS)· 2022-08-11
PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action Recognition Code
#42Hyperformer
92.9
Accuracy (CS)· 2022-11-17
Hypergraph Transformer for Skeleton-based Action Recognition Code
#43STGAT
92.8
Accuracy (CS)· 2022-08-18
Spatial Temporal Graph Attention Network for Skeleton-Based Action Recognition Code
#44TCA-GCN
92.8
Accuracy (CS)· 2022-05-31
Skeleton-based Action Recognition via Temporal-Channel Aggregation Code
#45TD-GCN
92.8
Accuracy (CS)
No paperCode
#46DSTSA-GCN
92.78
Accuracy (CS)· 2025-01-21
DSTSA-GCN: Advancing Skeleton-Based Gesture Recognition with Semantic-Aware Spatio-Temporal Topology Modeling Code
#47DSTSA-GCN
92.78
Accuracy (CS)· 2025-01-21
DSTSA-GCN: Advancing Skeleton-Based Gesture Recognition with Semantic-Aware Spatio-Temporal Topology Modeling Code
#48MSSTNet
92.6
Accuracy (CS)
No paperCode
#49ST-GCN++ [PYSKL, 3D Skeleton]
92.6
Accuracy (CS)· 2022-05-19
PYSKL: Towards Good Practices for Skeleton Action Recognition Code
#50TSMF (RGB + Pose)
92.5
Accuracy (CS)· Extra Data
No paperCode
#51CTR-GCN
92.4
Accuracy (CS)· 2021-07-26
Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition Code
#52MSAF (RGB+Pose)
92.24
Accuracy (CS)· 2020-12-13
MSAF: Multimodal Split Attention Fusion Code
#53EfficientGCN-B4
92.1
Accuracy (CS)· 2021-06-29
Constructing Stronger and Faster Baselines for Skeleton-based Action Recognition Code
#54STAR-Transformer (RGB + Pose)
92
Accuracy (CS)· 2022-10-14
STAR-Transformer: A Spatio-temporal Cross Attention Transformer for Human Action Recognition
#55DualHead-Net
92
Accuracy (CS)· 2021-08-10
Learning Multi-Granular Spatio-Temporal Graph Network for Skeleton-based Action Recognition Code
#56MMTM (RGB+Pose)
91.99
Accuracy (CS)· 2019-11-20
MMTM: Multimodal Transfer Module for CNN Fusion Code
#57FUSION (IR+Pose)
91.8
Accuracy (CS)· 2020-02-28
Infrared and 3D skeleton feature fusion for RGB-D action recognition Code
#58PoseMap (RGB+Pose)
91.7
Accuracy (CS)
No paper
#59B2C-AFM(RGB+Pose)
91.7
Accuracy (CS)
No paperCode
#60AngNet-JA + BA + JBA + VJBA
91.7
Accuracy (CS)· Extra Data· 2021-05-04
Fusing Higher-order Features in Graph Neural Networks for Skeleton-based Action Recognition Code
#61Skeletal GNN
91.6
Accuracy (CS)· Extra Data· 2021-08-16
Learning Skeletal Graph Neural Networks for Hard 3D Pose Estimation
#62SCK⊕
91.56
Accuracy (CS)· Extra Data· 2020-12-28
Tensor Representations for Action Recognition Code
#63DSTA-Net
91.5
Accuracy (CS)· 2020-07-07
Decoupled Spatial-Temporal Attention Network for Skeleton-Based Action Recognition Code
#64MS-G3D Net
91.5
Accuracy (CS)· Extra Data· 2020-03-31
Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition Code
#65Dynamic GCN
91.5
Accuracy (CS)· 2020-07-29
Dynamic GCN: Context-enriched Topology Learning for Skeleton-based Action Recognition Code
#66ST-GCN [PYSKL, 2D Skeleton]
91.4
Accuracy (CS)· 2022-05-19
PYSKL: Towards Good Practices for Skeleton Action Recognition Code
#67MS-AAGCN+TEM
91
Accuracy (CS)· 2020-03-19
Temporal Extension Module for Skeleton-Based Action Recognition
#68PA-ResGCN-B19
90.9
Accuracy (CS)· 2020-10-20
Stronger, Faster and More Explainable: A Graph Convolutional Baseline for Skeleton-based Action Recognition Code
#69EfficientGCN-B2
90.9
Accuracy (CS)· 2021-06-29
Constructing Stronger and Faster Baselines for Skeleton-based Action Recognition Code
#704s Shift-GCN
90.7
Accuracy (CS)· Extra Data
No paperCode
#71ST-GCN [PYSKL, 3D Skeleton]SOTA
90.7
Accuracy (CS)· 2018-01-23
Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition Code
#72CGCN
90.3
Accuracy (CS)· 2020-03-06
Unifying Graph Embedding Features with Graph Convolutional Networks for Skeleton-based Action Recognition
#73BAGCN
90.3
Accuracy (CS)· 2019-12-24
Focusing and Diffusion: Bidirectional Attentive Graph Convolutional Networks for Skeleton-based Action Recognition
#74FGCN-spatial+FGCN-motion
90.2
Accuracy (CS)· 2020-03-17
Feedback Graph Convolutional Network for Skeleton-based Action Recognition
#75Sym-GNN
90.1
Accuracy (CS)· 2019-10-05
Symbiotic Graph Neural Networks for 3D Skeleton-based Human Action Recognition and Motion Prediction
#76ST-GCN [Vanilla, 2D Skeleton]
90.1
Accuracy (CS)· 2018-01-23
Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition Code
#77Action Capsules
90
Accuracy (CS)· 2023-01-30
Action Capsules: Human Skeleton Action Recognition
#78MS-AAGCN
90
Accuracy (CS)· 2019-12-15
Skeleton-Based Action Recognition with Multi-Stream Adaptive Graph Convolutional Networks Code
#79DGNN
89.9
Accuracy (CS)
No paperCode
#80ST-TR-agcn
89.9
Accuracy (CS)· 2020-08-17
Skeleton-based Action Recognition via Spatial and Temporal Transformer Networks Code
#81EfficientGCN-B0
89.9
Accuracy (CS)· 2021-06-29
Constructing Stronger and Faster Baselines for Skeleton-based Action Recognition Code
#82Mix-Dimension
89.7
Accuracy (CS)· 2020-07-30
Mix Dimension in Poincaré Geometry for 3D Skeleton-based Action Recognition
#832s-SDGCN
89.58
Accuracy (CS)
No paper
#84TD-GDSCN
89.57
Accuracy (CS)
No paperCode
#85JB-AAGCN
89.4
Accuracy (CS)· 2019-12-15
Skeleton-Based Action Recognition with Multi-Stream Adaptive Graph Convolutional Networks Code
#86GCN-NAS
89.4
Accuracy (CS)· 2019-11-11
Learning Graph Convolutional Network for Skeleton-based Human Action Recognition by Neural Searching Code
#87VA-fusion (aug.)
89.4
Accuracy (CS)· 2018-04-20
View Adaptive Neural Networks for High Performance Skeleton-based Human Action Recognition Code
#88AGC-LSTM (Joint&Part)
89.2
Accuracy (CS)· 2019-02-25
An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition
#893s-HYSP
89.1
Accuracy (CS)· 2023-03-10
HYperbolic Self-Paced Learning for Self-Supervised Skeleton-based Action Representations Code
#90SLnL-rFA
89.1
Accuracy (CS)· 2018-11-10
Skeleton-Based Action Recognition with Synchronous Local and Non-local Spatio-temporal Learning and Frequency Attention
#91SGN
89
Accuracy (CS)· 2019-04-02
Semantics-Guided Neural Networks for Efficient Skeleton-Based Human Action Recognition Code
#92CoS-TR* (2-stream)
88.9
Accuracy (CS)· 2022-03-21
Continual Spatio-Temporal Graph Convolutional Networks Code
#933SCNN
88.6
Accuracy (CS)
No paper
#942s-NLGCN
88.5
Accuracy (CS)· 2018-05-20
Two-Stream Adaptive Graph Convolutional Networks for Skeleton-Based Action Recognition Code
#95AutoGCN
88.3
Accuracy (CS)· 2024-02-02
AutoGCN -- Towards Generic Human Activity Recognition with Neural Architecture Search Code
#96CoST-GCN* (2-stream)
88.3
Accuracy (CS)· 2022-03-21
Continual Spatio-Temporal Graph Convolutional Networks Code
#97PGCN-TCA
88
Accuracy (CS)
No paper
#98PB-GCN (Skeleton only)
87.5
Accuracy (CS)· 2018-09-13
Part-based Graph Convolutional Network for Action Recognition Code
#99GR-GCN
87.5
Accuracy (CS)· 2018-11-29
Optimized Skeleton-based Action Recognition via Sparsified Graph Regression
#100PB-GCN
87.5
Accuracy (CS)· 2018-09-13
Part-based Graph Convolutional Network for Action Recognition Code
#1013s RA-GCN
87.3
Accuracy (CS)· 2020-08-09
Richly Activated Graph Convolutional Network for Robust Skeleton-based Action Recognition Code
#102TS-SAN
87.2
Accuracy (CS)· 2019-12-18
Self-Attention Network for Skeleton-based Human Action Recognition
#1033s-USDRL (DSTE) This work
87.1
Accuracy (CS)· 2024-12-12
USDRL: Unified Skeleton-Based Dense Representation Learning with Multi-Grained Feature Decorrelation Code
#104AS-GCN
86.8
Accuracy (CS)· 2019-04-26
Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognition Code
#105RF-Action
86.8
Accuracy (CS)· 2019-09-20
Making the Invisible Visible: Action Recognition Through Walls and Occlusions
#106Dense IndRNN
86.7
Accuracy (CS)· 2019-10-11
Deep Independently Recurrent Neural Network (IndRNN)Code
#107Glimpse Clouds (RGB only)
86.6
Accuracy (CS)· 2018-02-22
Glimpse Clouds: Human Activity Recognition from Unstructured Feature Points Code
#108ST-GCN [Vanilla, 3D Skeleton]
86.6
Accuracy (CS)· 2018-01-23
Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition Code
#109HYSP
86.5
Accuracy (CS)· 2023-03-10
HYperbolic Self-Paced Learning for Self-Supervised Skeleton-based Action Representations Code
#110HCN
86.5
Accuracy (CS)· 2018-04-17
Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation Code
#111CoST-GCN*
86.3
Accuracy (CS)· 2022-03-21
Continual Spatio-Temporal Graph Convolutional Networks Code
#112CoS-TR*
86.3
Accuracy (CS)· 2022-03-21
Continual Spatio-Temporal Graph Convolutional Networks Code
#113Sem-GCN
86.2
Accuracy (CS)
No paper
#114ST-GCN
86
Accuracy (CS)· 2022-03-21
Continual Spatio-Temporal Graph Convolutional Networks Code
#115CoAGCN* (2-stream)
86
Accuracy (CS)· 2022-03-21
Continual Spatio-Temporal Graph Convolutional Networks Code
#1163s RA-GCN
85.9
Accuracy (CS)· 2019-05-16
Richly Activated Graph Convolutional Network for Action Recognition with Incomplete Skeletons Code
#1172s RA-GCN
85.8
Accuracy (CS)· 2019-05-16
Richly Activated Graph Convolutional Network for Action Recognition with Incomplete Skeletons Code
#118PeGCN
85.6
Accuracy (CS)· 2020-03-17
Predictively Encoded Graph Convolutional Network for Noise-Robust Skeleton-based Action Recognition Code
#1192D-3D-Softargmax (RGB only)
85.5
Accuracy (CS)· 2018-02-26
2D/3D Pose Estimation and Action Recognition using Multitask Deep Learning Code
#120GVFE + AS-GCN with DH-TCN
85.3
Accuracy (CS)· 2019-12-20
Vertex Feature Encoding and Hierarchical Temporal Modeling in a Spatial-Temporal Graph Convolutional Network for Action Recognition
#121PR-GCN
85.2
Accuracy (CS)· 2020-10-14
Pose Refinement Graph Convolutional Network for Skeleton-based Action Recognition
#122AR-GCN
85.1
Accuracy (CS)
No paper
#1233scale ResNet152SOTA
85
Accuracy (CS)· 2017-04-19
Skeleton based action recognition using translation-scale invariant image mapping and multi-scale deep cnn
#124SR-TSL
84.8
Accuracy (CS)· 2018-05-07
Skeleton-Based Action Recognition with Spatial Reasoning and Temporal Stack Learning
#1253s-ActCLR
84.3
Accuracy (CS)· 2023-03-20
Actionlet-Dependent Contrastive Learning for Unsupervised Skeleton-Based Action Recognition
#126RGB+Skeleton (cross-attention)
84.23
Accuracy (CS)
No paper
#127CoAGCN*
84.1
Accuracy (CS)· 2022-03-21
Continual Spatio-Temporal Graph Convolutional Networks Code
#128ST-GCN-jpd
83.36
Accuracy (CS)· 2019-06-24
A Comparative Review of Recent Kinect-based Action Recognition Algorithms Code
#129CNN+Motion+Trans
83.2
Accuracy (CS)· 2017-04-25
Skeleton-based Action Recognition with Convolutional Neural Networks Code
#130IndRNN (with jpd)
83
Accuracy (CS)· 2019-06-24
A Comparative Review of Recent Kinect-based Action Recognition Algorithms Code
#131FO-GASTM
82.83
Accuracy (CS)
No paper
#132MANs (DenseNet-161)
82.67
Accuracy (CS)· 2018-04-23
Memory Attention Networks for Skeleton-based Action Recognition Code
#133DM-3DCNN
82
Accuracy (CS)
No paperCode
#134Bayesian GC-LSTM
81.8
Accuracy (CS)
No paper
#135Ind-RNN
81.8
Accuracy (CS)· 2018-03-13
Independently Recurrent Neural Network (IndRNN): Building A Longer and Deeper RNN Code
#136ST-GCN
81.5
Accuracy (CS)· 2018-01-23
Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition Code
#137HPM_RGB+HPM_3D+Traj
80.9
Accuracy (CS)· 2017-07-04
Learning Human Pose Models from Synthesized Data for Robust RGB-D Action Recognition
#138ARRN-LSTM
80.7
Accuracy (CS)· 2018-05-07
Relational Network for Skeleton-Based Action Recognition
#139EleAtt-GRU (aug.)
80.7
Accuracy (CS)· 2019-09-03
EleAtt-RNN: Adding Attentiveness to Neurons in Recurrent Neural Networks
#140Skeleton Visualization
80
Accuracy (CS)
No paper
#141EleAtt-GRU
79.8
Accuracy (CS)· 2018-07-12
Adding Attentiveness to the Neurons in Recurrent Neural Networks
#142Clips+CNN+MTLNSOTA
79.6
Accuracy (CS)· 2017-03-09
A New Representation of Skeleton Sequences for 3D Action Recognition
#143F2CSkeleton
79.6
Accuracy (CS)· 2018-05-30
A Fine-to-Coarse Convolutional Neural Network for 3D Human Action Recognition
#144VA-LSTM
79.2
Accuracy (CS)· 2017-03-24
View Adaptive Recurrent Neural Networks for High Performance Human Action Recognition from Skeleton Data Code
#145Skelemotion + Yang et al. (Skeleton only)
76.5
Accuracy (CS)· 2019-07-30
SkeleMotion: A New Representation of Skeleton Joint Sequences Based on Motion Information for 3D Action Recognition Code
#146Skelemotion + Yang et al.
76.5
Accuracy (CS)· 2019-07-30
SkeleMotion: A New Representation of Skeleton Joint Sequences Based on Motion Information for 3D Action Recognition Code
#147GCA-LSTM
76.1
Accuracy (CS)
No paper
#148SkeletonNet
75.9
Accuracy (CS)
No paper
#149DSSCA-SSLM (RGB only)SOTA
74.9
Accuracy (CS)· 2016-03-23
Deep Multimodal Feature Analysis for Action Recognition in RGB+D Videos
#150URNN-2L-T
74.6
Accuracy (CS)
No paper
#151Ensemble TS-LSTM v2
74.6
Accuracy (CS)
No paperCode
#152TCN
74.3
Accuracy (CS)· 2017-04-14
Interpretable 3D Human Action Analysis with Temporal Convolutional Networks Code
#153STA-LSTM
73.4
Accuracy (CS)· 2016-11-18
An End-to-End Spatio-Temporal Attention Model for Human Action Recognition from Skeleton Data
#154TSRJI (Late Fusion) + HCN
73.3
Accuracy (CS)· 2019-09-11
Skeleton Image Representation for 3D Action Recognition based on Tree Structure and Reference Joints Code
#155Two-Stream RNN
71.3
Accuracy (CS)· 2017-04-09
Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks
#156Spatio-Temporal LSTM
69.2
Accuracy (CS)· 2016-07-24
Spatio-Temporal LSTM with Trust Gates for 3D Human Action Recognition
#157Part-aware LSTM
62.93
Accuracy (CS)· 2016-04-11
NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis Code
#158ST-LSTM
61.7
Accuracy (CS)· 2016-07-24
Spatio-Temporal LSTM with Trust Gates for 3D Human Action Recognition
#159Deep LSTM
60.7
Accuracy (CS)· 2016-04-11
NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis Code
#160Dynamic Skeletons
60.2
Accuracy (CS)
No paper
#161H-RNN
59.1
Accuracy (CS)
No paper
#162Lie Group
50.1
Accuracy (CS)
No paperCode
#163Skeleton Quads
38.6
Accuracy (CS)
No paper