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/VidTr-L

VidTr-L

Reported on 9 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.

Computer Vision3 results

  • VideoonKinetics-700
    Top-1 Accuracy· 2021-04-23
    70.2
    best: 85.9 (InternVideo2-6B)
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746
  • VideoonKinetics-700
    Top-5 Accuracy· 2021-04-23
    89
    best: 96.7 (UMT-L (ViT-L/16))
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746
  • VideoonCharades
    MAP· 2021-04-23
    43.5
    best: 66.3 (TokenLearner)
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746

Robots3 results

  • Activity RecognitiononHMDB-51
    Average accuracy of 3 splits· 2021-04-23
    74.4
    best: 88.7 (VideoMAE V2-g)
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746
  • Activity RecognitiononSomething-Something V2
    Top-1 Accuracy· 2021-04-23
    60.2
    best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746
  • Activity RecognitiononUCF101
    3-fold Accuracy· 2021-04-23
    96.7
    best: 99.7 (FTP-UniFormerV2-L/14)
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746

Time Series3 results

  • Action RecognitiononHMDB-51
    Average accuracy of 3 splits· 2021-04-23
    74.4
    best: 88.7 (VideoMAE V2-g)
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746
  • Action RecognitiononSomething-Something V2
    Top-1 Accuracy· 2021-04-23
    60.2
    best: 77.3 (MVD (Kinetics400 pretrain, ViT-H, 16 frame))
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746
  • Action RecognitiononUCF101
    3-fold Accuracy· 2021-04-23
    96.7
    best: 99.7 (FTP-UniFormerV2-L/14)
    VidTr: Video Transformer Without ConvolutionsarXiv:2104.11746