TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/ProtoGCN

ProtoGCN

Reported on 56 benchmarks across 8 tasks · 1 paper · 12 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision21 results

  • VideoonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • VideoonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Temporal Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Temporal Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • VideoonNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • VideoonKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • VideoonNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 94.3 (Hulk(Finetune, ViT-L))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • VideoonNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • VideoonNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Temporal Action LocalizationonNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Temporal Action LocalizationonKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Temporal Action LocalizationonNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 94.3 (Hulk(Finetune, ViT-L))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Temporal Action LocalizationonNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Temporal Action LocalizationonNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action LocalizationonNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action LocalizationonKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action LocalizationonNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 94.3 (Hulk(Finetune, ViT-L))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action LocalizationonNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action LocalizationonNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941

Time Series14 results

  • Action DetectiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action DetectiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action DetectiononNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action DetectiononKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action DetectiononNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 94.3 (Hulk(Finetune, ViT-L))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action DetectiononNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action DetectiononNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    best: 96.7 (DSCNet (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    best: 95.6 (DSCNet (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action RecognitiononNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action RecognitiononKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action RecognitiononNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 97.4 (DSCNet (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action RecognitiononNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 99.6 (PoseC3D (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Action RecognitiononNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941

Methodology7 results

  • Zero-Shot LearningonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Zero-Shot LearningonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Zero-Shot LearningonNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Zero-Shot LearningonKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Zero-Shot LearningonNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 94.3 (Hulk(Finetune, ViT-L))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Zero-Shot LearningonNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Zero-Shot LearningonNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941

Natural Language Processing7 results

  • 3D Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • 3D Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    SOTA
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • 3D Action RecognitiononNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • 3D Action RecognitiononKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • 3D Action RecognitiononNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 94.3 (Hulk(Finetune, ViT-L))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • 3D Action RecognitiononNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • 3D Action RecognitiononNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941

Robots7 results

  • Activity RecognitiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2024-11-28
    92.2
    best: 96.7 (DSCNet (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Activity RecognitiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2024-11-28
    90.9
    best: 95.6 (DSCNet (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Activity RecognitiononNTU RGB+D 120
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Activity RecognitiononKinetics-Skeleton dataset
    Accuracy· uses extra data· 2024-11-28
    51.9
    best: 52.3 (Structured Keypoint Pooling (PPNv2 skeletons+objects))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Activity RecognitiononNTU RGB+D
    Accuracy (CS)· 2024-11-28
    93.8
    best: 97.4 (DSCNet (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Activity RecognitiononNTU RGB+D
    Accuracy (CV)· 2024-11-28
    97.8
    best: 99.6 (PoseC3D (RGB + Pose))
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941
  • Activity RecognitiononNTU RGB+D
    Ensembled Modalities· 2024-11-28
    6
    Revealing Key Details to See Differences: A Novel Prototypical Perspective for Skeleton-based Action RecognitionarXiv:2411.18941