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/MTV-H (WTS 60M)

MTV-H (WTS 60M)

Reported on 8 benchmarks across 1 task · 1 paper · 7 SOTA

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

Computer Vision8 results

  • VideoonKinetics-700
    Top-1 Accuracy· uses extra data· 2022-01-12
    83.4
    best: 85.9 (InternVideo2-6B)
    SOTA
    Multiview Transformers for Video RecognitionarXiv:2201.04288
  • VideoonKinetics-700
    Top-5 Accuracy· uses extra data· 2022-01-12
    96.2
    best: 96.7 (UMT-L (ViT-L/16))
    SOTA
    Multiview Transformers for Video RecognitionarXiv:2201.04288
  • VideoonMiT
    Top 1 Accuracy· uses extra data· 2022-01-12
    47.2
    best: 53.1 (OmniVec2)
    SOTA
    Multiview Transformers for Video RecognitionarXiv:2201.04288
  • VideoonMiT
    Top 5 Accuracy· uses extra data· 2022-01-12
    75.7
    best: 78.2 (UMT-L (ViT-L/16))
    SOTA
    Multiview Transformers for Video RecognitionarXiv:2201.04288
  • VideoonKinetics-400
    Acc@1· 2022-01-12
    89.9
    best: 93.6 (OmniVec2)
    SOTA
    Multiview Transformers for Video RecognitionarXiv:2201.04288
  • VideoonKinetics-400
    Acc@5· 2022-01-12
    98.3
    best: 98.9 (TubeViT-H (ImageNet-1k))
    SOTA
    Multiview Transformers for Video RecognitionarXiv:2201.04288
  • VideoonKinetics-600
    Top-5 Accuracy· uses extra data· 2022-01-12
    98.5
    best: 98.9 (TubeVit-H)
    SOTA
    Multiview Transformers for Video RecognitionarXiv:2201.04288
  • VideoonKinetics-600
    Top-1 Accuracy· uses extra data· 2022-01-12
    90.3
    best: 91.9 (InternVideo2-6B)
    Multiview Transformers for Video RecognitionarXiv:2201.04288