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/MeMViT-24

MeMViT-24

Reported on 13 benchmarks across 5 tasks · 1 paper · 5 SOTA

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

Robots5 results

  • Activity RecognitiononEPIC-KITCHENS-100
    Recall@5· uses extra data· 2022-01-20
    17.7
    best: 27.6 (PlausiVL)
    SOTA
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Activity RecognitiononEPIC-KITCHENS-100
    Action@1· uses extra data· 2022-01-20
    48.4
    best: 58.3 (LLaVAction)
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Activity RecognitiononEPIC-KITCHENS-100
    Noun@1· uses extra data· 2022-01-20
    60.3
    best: 69 (LLaVAction)
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Activity RecognitiononEPIC-KITCHENS-100
    Verb@1· uses extra data· 2022-01-20
    71.4
    best: 76.2 (TIM)
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Activity RecognitiononAVA v2.2
    mAP· uses extra data· 2022-01-20
    35.4
    best: 45.1 (LART (Hiera-H, K700 PT+FT))
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383

Time Series5 results

  • Action RecognitiononEPIC-KITCHENS-100
    Recall@5· uses extra data· 2022-01-20
    17.7
    best: 27.6 (PlausiVL)
    SOTA
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Action RecognitiononEPIC-KITCHENS-100
    Action@1· uses extra data· 2022-01-20
    48.4
    best: 58.3 (LLaVAction)
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Action RecognitiononEPIC-KITCHENS-100
    Noun@1· uses extra data· 2022-01-20
    60.3
    best: 69 (LLaVAction)
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Action RecognitiononEPIC-KITCHENS-100
    Verb@1· uses extra data· 2022-01-20
    71.4
    best: 76.2 (TIM)
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Action RecognitiononAVA v2.2
    mAP· uses extra data· 2022-01-20
    35.4
    best: 45.1 (LART (Hiera-H, K700 PT+FT))
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383

Computer Vision2 results

  • Action AnticipationonEPIC-KITCHENS-100
    Recall@5· uses extra data· 2022-01-20
    17.7
    best: 27.6 (PlausiVL)
    SOTA
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383
  • Action Recognition In VideosonEPIC-KITCHENS-100
    Recall@5· uses extra data· 2022-01-20
    17.7
    best: 27.6 (PlausiVL)
    SOTA
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383

Knowledge Base1 result

  • 2D Human Pose EstimationonEPIC-KITCHENS-100
    Recall@5· uses extra data· 2022-01-20
    17.7
    best: 27.6 (PlausiVL)
    SOTA
    MeMViT: Memory-Augmented Multiscale Vision Transformer for Efficient Long-Term Video RecognitionarXiv:2201.08383