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/MAR (50% mask, ViT-B, 16x4)

MAR (50% mask, ViT-B, 16x4)

Reported on 8 benchmarks across 3 tasks · 1 paper

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

Robots3 results

  • Activity RecognitiononSomething-Something V2
    Parameters· 2022-07-24
    94
    best: 2131 (InternVideo2-6B)
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660
  • Activity RecognitiononSomething-Something V2
    Top-1 Accuracy· 2022-07-24
    71
    best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660
  • Activity RecognitiononSomething-Something V2
    Top-5 Accuracy· 2022-07-24
    92.8
    best: 96.3 (DejaVid)
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660

Time Series3 results

  • Action RecognitiononSomething-Something V2
    Parameters· 2022-07-24
    94
    best: 2131 (InternVideo2-6B)
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660
  • Action RecognitiononSomething-Something V2
    Top-1 Accuracy· 2022-07-24
    71
    best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660
  • Action RecognitiononSomething-Something V2
    Top-5 Accuracy· 2022-07-24
    92.8
    best: 96.3 (DejaVid)
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660

Computer Vision2 results

  • VideoonKinetics-400
    Acc@1· 2022-07-24
    81
    best: 93.6 (OmniVec2)
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660
  • VideoonKinetics-400
    Acc@5· 2022-07-24
    94.4
    best: 98.9 (TubeViT-H (ImageNet-1k))
    MAR: Masked Autoencoders for Efficient Action RecognitionarXiv:2207.11660