Metric Learning On In Shop 1

Metrics

R@1

Results

Performance results of various models on this benchmark

		Paper Title
Unicom+ViT-L@336px	96.7	Unicom: Universal and Compact Representation Learning for Image Retrieval
STIR	95	STIR: Siamese Transformer for Image Retrieval Postprocessing
MGA	94.3	Fashion Image Retrieval with Multi-Granular Alignment
Hyp-ViT	92.5	Hyperbolic Vision Transformers: Combining Improvements in Metric Learning
Hyp-DINO	92.4	Hyperbolic Vision Transformers: Combining Improvements in Metric Learning
CCL (ResNet-50)	92.31	Center Contrastive Loss for Metric Learning
Gradient Surgery	92.21	Dissecting the impact of different loss functions with gradient surgery
ResNet-50 + Metrix	92.2	It Takes Two to Tango: Mixup for Deep Metric Learning
EfficientDML-VPTSP-G/512	92.1	Learning Semantic Proxies from Visual Prompts for Parameter-Efficient Fine-Tuning in Deep Metric Learning
ViT-Triplet	92.1	STIR: Siamese Transformer for Image Retrieval Postprocessing
NED	91.3	Calibrated neighborhood aware confidence measure for deep metric learning
ResNet-50 + ProxyNCA++	90.9	ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis
ResNet-50 + Cross-Entropy	90.6	A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses
SCT(512)	90	Hard negative examples are hard, but useful
EPSHN(512)	87.8	Improved Embeddings with Easy Positive Triplet Mining

0 of 15 row(s) selected.

Metric Learning On In Shop 1

Metrics

R@1

Results

Performance results of various models on this benchmark

		Paper Title
Unicom+ViT-L@336px	96.7	Unicom: Universal and Compact Representation Learning for Image Retrieval
STIR	95	STIR: Siamese Transformer for Image Retrieval Postprocessing
MGA	94.3	Fashion Image Retrieval with Multi-Granular Alignment
Hyp-ViT	92.5	Hyperbolic Vision Transformers: Combining Improvements in Metric Learning
Hyp-DINO	92.4	Hyperbolic Vision Transformers: Combining Improvements in Metric Learning
CCL (ResNet-50)	92.31	Center Contrastive Loss for Metric Learning
Gradient Surgery	92.21	Dissecting the impact of different loss functions with gradient surgery
ResNet-50 + Metrix	92.2	It Takes Two to Tango: Mixup for Deep Metric Learning
EfficientDML-VPTSP-G/512	92.1	Learning Semantic Proxies from Visual Prompts for Parameter-Efficient Fine-Tuning in Deep Metric Learning
ViT-Triplet	92.1	STIR: Siamese Transformer for Image Retrieval Postprocessing
NED	91.3	Calibrated neighborhood aware confidence measure for deep metric learning
ResNet-50 + ProxyNCA++	90.9	ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis
ResNet-50 + Cross-Entropy	90.6	A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses
SCT(512)	90	Hard negative examples are hard, but useful
EPSHN(512)	87.8	Improved Embeddings with Easy Positive Triplet Mining

0 of 15 row(s) selected.

Metric Learning On In Shop 1 | SOTA | HyperAI