Metric Learning On Cars196

评估指标

R@1

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
Unicom+ViT-L@336px	98.2	Unicom: Universal and Compact Representation Learning for Image Retrieval
Hyp-DINO 8x8	92.8	Hyperbolic Vision Transformers: Combining Improvements in Metric Learning
NED	91.5	Calibrated neighborhood aware confidence measure for deep metric learning	-
ResNet-50 + AVSL	91.5	Attributable Visual Similarity Learning
ResNet-50 + Intra-Batch (ensemble of 5)	91.5	Learning Intra-Batch Connections for Deep Metric Learning
EfficientDML-VPTSP-G/512	91.2	Learning Semantic Proxies from Visual Prompts for Parameter-Efficient Fine-Tuning in Deep Metric Learning
CCL (ResNet-50)	91.02	Center Contrastive Loss for Metric Learning	-
ResNet50 + Language	90.2	Integrating Language Guidance into Vision-based Deep Metric Learning
ResNet-50 + Metrix	89.6	It Takes Two to Tango: Mixup for Deep Metric Learning
Recall@k Surrogate loss (ViT-B/16)	89.5	Recall@k Surrogate Loss with Large Batches and Similarity Mixup
ResNet50 + S2SD	89.5	S2SD: Simultaneous Similarity-based Self-Distillation for Deep Metric Learning
ResNet-50 + Cross-Entropy	89.3	A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses
Hyp-DINO	89.2	Hyperbolic Vision Transformers: Combining Improvements in Metric Learning
ResNet50 + NIR	89.1	Non-isotropy Regularization for Proxy-based Deep Metric Learning
Margin + DAS	88.34	DAS: Densely-Anchored Sampling for Deep Metric Learning
BN-Inception + Proxy-Anchor	88.3	Proxy Anchor Loss for Deep Metric Learning
Recall@k Surrogate loss (ResNet-50)	88.3	Recall@k Surrogate Loss with Large Batches and Similarity Mixup
ResNet-50 + Intra-Batch	88.1	Learning Intra-Batch Connections for Deep Metric Learning
ABE + HORDE	88.0	Metric Learning With HORDE: High-Order Regularizer for Deep Embeddings
MS + SEC + DAS	87.8	DAS: Densely-Anchored Sampling for Deep Metric Learning

0 of 36 row(s) selected.