ResNet-50 + Cross-Entropy

Reported on 4 benchmarks across 1 task · 1 paper · 4 SOTA

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

Methodology4 results

Metric LearningonCARS196
R@1· uses extra data· 2020-03-19
89.3
best: 98.2 (Unicom+ViT-L@336px)
SOTA
A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses arXiv:2003.08983
Metric Learningon CUB-200-2011
R@1· uses extra data· 2020-03-19
69.2
best: 90.1 (Unicom+ViT-L@336px)
SOTA
A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses arXiv:2003.08983
Metric LearningonIn-Shop
R@1· uses extra data· 2020-03-19
90.6
best: 96.7 (Unicom+ViT-L@336px)
SOTA
A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses arXiv:2003.08983
Metric LearningonStanford Online Products
R@1· uses extra data· 2020-03-19
81.1
best: 91.2 (Unicom+ViT-L@336px)
SOTA
A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses arXiv:2003.08983