VideoMS (ViT-B)

Reported on 4 benchmarks across 2 tasks · 1 paper

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

Robots2 results

Activity RecognitiononUCF101
3-fold Accuracy· 2022-11-19
93.4
best: 99.7 (FTP-UniFormerV2-L/14)
EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens arXiv:2211.10636
Activity RecognitiononHMDB51
Top-1 Accuracy· 2022-11-19
65.8
best: 79.7 (MVD (ViT-B))
EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens arXiv:2211.10636

Time Series2 results

Action RecognitiononUCF101
3-fold Accuracy· 2022-11-19
93.4
best: 99.7 (FTP-UniFormerV2-L/14)
EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens arXiv:2211.10636
Action RecognitiononHMDB51
Top-1 Accuracy· 2022-11-19
65.8
best: 79.7 (MVD (ViT-B))
EVEREST: Efficient Masked Video Autoencoder by Removing Redundant Spatiotemporal Tokens arXiv:2211.10636