HyperAIHyperAI超神经

Command Palette

Search for a command to run...

算力平台

Skeleton Based Action Recognition On Ntu Rgbd

评估指标

Accuracy (CS)
Accuracy (CV)

评测结果

各个模型在此基准测试上的表现结果

Paper Title
Lit-llama9598.4LLMs are Good Action Recognizers
Hulk(Finetune, ViT-L)94.3-Hulk: A Universal Knowledge Translator for Human-Centric Tasks
PoseC3D [3D Heatmap]94.197.1Revisiting Skeleton-based Action Recognition
Hulk(Finetune, ViT-B)94-Hulk: A Universal Knowledge Translator for Human-Centric Tasks
MaskCLR93.997.3MaskCLR: Attention-Guided Contrastive Learning for Robust Action Representation Learning
ProtoGCN93.897.8Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action Recognition
Shap-Mix93.797.1Shap-Mix: Shapley Value Guided Mixing for Long-Tailed Skeleton Based Action Recognition
JMDA (based on Skeleton MixFormer)93.797.2Joint Mixing Data Augmentation for Skeleton-based Action Recognition
DeGCN93.697.4DeGCN: Deformable Graph Convolutional Networks for Skeleton-Based Action Recognition
MSA-GCN93.697.4MSA-GCN: Exploiting Multi-Scale Temporal Dynamics With Adaptive Graph Convolution for Skeleton-Based Action Recognition
SkateFormer93.597.8SkateFormer: Skeletal-Temporal Transformer for Human Action Recognition
LA-GCN93.597.2Language Knowledge-Assisted Representation Learning for Skeleton-Based Action Recognition
MMCL93.597.4Multi-Modality Co-Learning for Efficient Skeleton-based Action Recognition
HD-GCN93.497.2Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition
STEP-CATFormer93.297.3STEP CATFormer: Spatial-Temporal Effective Body-Part Cross Attention Transformer for Skeleton-based Action Recognition
DG-STGCN93.297.5DG-STGCN: Dynamic Spatial-Temporal Modeling for Skeleton-based Action Recognition
SkeletonGCL (based on CTR-GCN)93.197.0Graph Contrastive Learning for Skeleton-based Action Recognition
MAMP93.197.5Masked Motion Predictors are Strong 3D Action Representation Learners
BlockGCN93.197.0BlockGCN: Redefine Topology Awareness for Skeleton-Based Action Recognition
MotionBert (finetune)93.097.2MotionBERT: A Unified Perspective on Learning Human Motion Representations
0 of 133 row(s) selected.