XKD-Modality-Agnostic (ViT-B/112/16)

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· 2022-11-25
93.4
best: 99.7 (FTP-UniFormerV2-L/14)
XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning arXiv:2211.13929
Activity RecognitiononHMDB51
Top-1 Accuracy· 2022-11-25
65.9
best: 79.7 (MVD (ViT-B))
XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning arXiv:2211.13929

Time Series2 results

Action RecognitiononUCF101
3-fold Accuracy· 2022-11-25
93.4
best: 99.7 (FTP-UniFormerV2-L/14)
XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning arXiv:2211.13929
Action RecognitiononHMDB51
Top-1 Accuracy· 2022-11-25
65.9
best: 79.7 (MVD (ViT-B))
XKD: Cross-modal Knowledge Distillation with Domain Alignment for Video Representation Learning arXiv:2211.13929