AI Paper Atlas

Research map / Benchmarks / NTU120-XSet

NTU120-XSet leaderboard

12 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on NTU120-XSet. Best accuracy: 96.4 (RGBPose-Conv3D, CVPR 2022).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

NTU120-XSet · accuracy

12 papers, each with its best reported number for its own method; higher is better.

#accuracy ↑MethodPaperTopic
196.4RGBPose-Conv3D
SlowOnly, R50
Revisiting Skeleton-Based Action Recognition
CVPR 2022 · Table: Multi-Modality Action R
Recognition and long-tail
294.73Mformer Taylor skeleton
3Mformer
Taylor Videos for Action Recognition
ICML 2024 · Table 4
Action recognition
393.83Mformer (rank-pool)
Higher-order Transformer
3Mformer: Multi-Order Multi-Mode Transformer for Skeletal Action Recognition
CVPR 2023 · Table: Experimental results on
Knowledge and temporal graphs
492.6OursRethinking Masked Data Reconstruction Pretraining for Strong 3D Action Representation Learning
AAAI 2025 · Table 2
Self-supervised and contrastive
591.9MaskCLR
Transformer
MaskCLR: Attention-Guided Contrastive Learning for Robust Action Representation Learning
CVPR 2024 · Table 1
Recognition and long-tail
691.7JT-GraphFormer (4s)Spatio-Temporal Fusion for Human Action Recognition via Joint Trajectory Graph
AAAI 2024 · Table 5
Action recognition
791.5VA-ARVA-AR: Learning Velocity-Aware Action Representations with Mixture of Window Attention
AAAI 2025 · Table 1
Action recognition
891.4DG-STGCN(A0.66)
DG-STGCN
Are Spatial-Temporal Graph Convolution Networks for Human Action Recognition Over-Parameterized?
CVPR 2025 · Table 3
CNN design
991.4Ours (3D deformable transformer, R+P, no pre-training)Cross-Modal Learning with 3D Deformable Attention for Action Recognition
ICCV 2023 · Table 3
Action recognition
1091.1DS-GCNDynamic Semantic-Based Spatial Graph Convolution Network for Skeleton-Based Human Action Recognition
AAAI 2024 · Table 6
Knowledge and temporal graphs
2 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.