M3Video

Reported on 4 benchmarks across 2 tasks · 1 paper · 2 SOTA

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

Robots2 results

Activity RecognitiononHMDB51
Top-1 Accuracy· 2022-10-12
78
best: 79.7 (MVD (ViT-B))
SOTA
Masked Motion Encoding for Self-Supervised Video Representation Learning arXiv:2210.06096
Activity RecognitiononUCF101
3-fold Accuracy· 2022-10-12
96.5
best: 99.7 (FTP-UniFormerV2-L/14)
Masked Motion Encoding for Self-Supervised Video Representation Learning arXiv:2210.06096

Time Series2 results

Action RecognitiononHMDB51
Top-1 Accuracy· 2022-10-12
78
best: 79.7 (MVD (ViT-B))
SOTA
Masked Motion Encoding for Self-Supervised Video Representation Learning arXiv:2210.06096
Action RecognitiononUCF101
3-fold Accuracy· 2022-10-12
96.5
best: 99.7 (FTP-UniFormerV2-L/14)
Masked Motion Encoding for Self-Supervised Video Representation Learning arXiv:2210.06096