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/PSUMNet

PSUMNet

Reported on 32 benchmarks across 8 tasks · 1 paper

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

Computer Vision12 results

  • VideoonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 92.2 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • VideoonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 90.9 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • VideoonNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 94.3 (Hulk(Finetune, ViT-L))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • VideoonNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Temporal Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 92.2 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Temporal Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 90.9 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Temporal Action LocalizationonNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 94.3 (Hulk(Finetune, ViT-L))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Temporal Action LocalizationonNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 92.2 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action LocalizationonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 90.9 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action LocalizationonNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 94.3 (Hulk(Finetune, ViT-L))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action LocalizationonNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775

Time Series8 results

  • Action DetectiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 92.2 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action DetectiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 90.9 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action DetectiononNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 94.3 (Hulk(Finetune, ViT-L))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action DetectiononNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 96.7 (DSCNet (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 95.6 (DSCNet (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action RecognitiononNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 97.4 (DSCNet (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Action RecognitiononNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 99.6 (PoseC3D (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775

Methodology4 results

  • Zero-Shot LearningonNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 92.2 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Zero-Shot LearningonNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 90.9 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Zero-Shot LearningonNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 94.3 (Hulk(Finetune, ViT-L))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Zero-Shot LearningonNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775

Robots4 results

  • Activity RecognitiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 96.7 (DSCNet (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Activity RecognitiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 95.6 (DSCNet (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Activity RecognitiononNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 97.4 (DSCNet (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • Activity RecognitiononNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 99.6 (PoseC3D (RGB + Pose))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775

Natural Language Processing4 results

  • 3D Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Setup)· 2022-08-11
    90.6
    best: 92.2 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • 3D Action RecognitiononNTU RGB+D 120
    Accuracy (Cross-Subject)· 2022-08-11
    89.4
    best: 90.9 (ProtoGCN)
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • 3D Action RecognitiononNTU RGB+D
    Accuracy (CS)· 2022-08-11
    92.9
    best: 94.3 (Hulk(Finetune, ViT-L))
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775
  • 3D Action RecognitiononNTU RGB+D
    Accuracy (CV)· 2022-08-11
    96.7
    best: 98.3 (ST-GCN [PYSKL, 2D Skeleton])
    PSUMNet: Unified Modality Part Streams are All You Need for Efficient Pose-based Action RecognitionarXiv:2208.05775