VideoMAE(no extra data)

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· 2022-03-23
91.3
best: 99.7 (FTP-UniFormerV2-L/14)
VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training arXiv:2203.12602
Activity RecognitiononHMDB51
Top-1 Accuracy· 2022-03-23
62.6
best: 79.7 (MVD (ViT-B))
VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training arXiv:2203.12602

Time Series2 results

Action RecognitiononUCF101
3-fold Accuracy· 2022-03-23
91.3
best: 99.7 (FTP-UniFormerV2-L/14)
VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training arXiv:2203.12602
Action RecognitiononHMDB51
Top-1 Accuracy· 2022-03-23
62.6
best: 79.7 (MVD (ViT-B))
VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training arXiv:2203.12602