STIR

Reported on 2 benchmarks across 1 task · 1 paper

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

Methodology2 results

Metric LearningonIn-Shop
R@1· 2023-04-26
95
best: 96.7 (Unicom+ViT-L@336px)
STIR: Siamese Transformer for Image Retrieval Postprocessing arXiv:2304.13393
Metric LearningonStanford Online Products
R@1· 2023-04-26
88.3
best: 91.2 (Unicom+ViT-L@336px)
STIR: Siamese Transformer for Image Retrieval Postprocessing arXiv:2304.13393