Frozen Backbone, SwinV2-G-ext22K (Video-Swin)

Reported on 3 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.

Robots1 result

Activity RecognitiononKinetics-400
Top-1 Accuracy· 2022-11-03
81.7
best: 86.5 (Florence)
Could Giant Pretrained Image Models Extract Universal Representations?arXiv:2211.02043

Time Series1 result

Action RecognitiononKinetics-400
Top-1 Accuracy· 2022-11-03
81.7
best: 86.5 (Florence)
Could Giant Pretrained Image Models Extract Universal Representations?arXiv:2211.02043

Computer Vision1 result

Action Recognition In VideosonKinetics-400
Top-1 Accuracy· 2022-11-03
81.7
best: 86.5 (Florence)
Could Giant Pretrained Image Models Extract Universal Representations?arXiv:2211.02043