ROADMAP (DeiT-S)

Reported on 5 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.

Computer Vision4 results

Image RetrievaloniNaturalist
R@1· 2021-10-01
73.6
best: 88.9 (Unicom+ViT-L@336px)
Robust and Decomposable Average Precision for Image Retrieval arXiv:2110.01445
Image RetrievaloniNaturalist
R@16· 2021-10-01
93.1
best: 95.9 (Recall@k Surrogate loss (ViT-B/16))
Robust and Decomposable Average Precision for Image Retrieval arXiv:2110.01445
Image RetrievaloniNaturalist
R@32· 2021-10-01
95.2
best: 97.2 (Recall@k Surrogate loss (ViT-B/16))
Robust and Decomposable Average Precision for Image Retrieval arXiv:2110.01445
Image RetrievaloniNaturalist
R@5· 2021-10-01
86.2
best: 92.1 (Recall@k Surrogate loss (ViT-B/16))
Robust and Decomposable Average Precision for Image Retrieval arXiv:2110.01445

Methodology1 result

Metric LearningonStanford Online Products
R@1· 2021-10-01
86
best: 91.2 (Unicom+ViT-L@336px)
Robust and Decomposable Average Precision for Image Retrieval arXiv:2110.01445