| Unicom: Universal and Compact Representation Learning for Image Retrieval | ✓ Link | 91.2 | Unicom+ViT-L@336px | 2023-04-12 |
| STIR: Siamese Transformer for Image Retrieval Postprocessing | ✓ Link | 88.3 | STIR | 2023-04-26 |
| Recall@k Surrogate Loss with Large Batches and Similarity Mixup | ✓ Link | 88.0 | Recall@k Surrogate Loss (ViT-B/16) | 2021-08-25 |
| STIR: Siamese Transformer for Image Retrieval Postprocessing | ✓ Link | 86.5 | ViT-Triplet | 2023-04-26 |
| Robust and Decomposable Average Precision for Image Retrieval | ✓ Link | 86.0 | ROADMAP (DeiT-S) | 2021-10-01 |
| Hyperbolic Vision Transformers: Combining Improvements in Metric Learning | ✓ Link | 85.9 | Hyp-ViT | 2022-03-21 |
| Hyperbolic Vision Transformers: Combining Improvements in Metric Learning | ✓ Link | 85.1 | Hyp-DINO | 2022-03-21 |
| Recall@k Surrogate Loss with Large Batches and Similarity Mixup | ✓ Link | 85.1 | Recall@k Surrogate Loss (ViT-B/32) | 2021-08-25 |
| Robust and Decomposable Average Precision for Image Retrieval | ✓ Link | 83.1 | ROADMAP (ResNet-50) | 2021-10-01 |
| Center Contrastive Loss for Metric Learning | | 83.10 | CCL (ResNet-50) | 2023-08-01 |
| Recall@k Surrogate Loss with Large Batches and Similarity Mixup | ✓ Link | 82.7 | Recall@k Surrogate Loss (ResNet-50) | 2021-08-25 |
| Dissecting the impact of different loss functions with gradient surgery | | 82.3 | Gradient Surgery | 2022-01-27 |
| Hierarchical Average Precision Training for Pertinent Image Retrieval | ✓ Link | 81.8 | HAPPIER_F | 2022-07-05 |
| Hard negative examples are hard, but useful | ✓ Link | 81.6 | SCT(512) | 2020-07-24 |
| It Takes Two to Tango: Mixup for Deep Metric Learning | ✓ Link | 81.3 | ResNet-50 + Metrix | 2021-06-09 |
| Integrating Language Guidance into Vision-based Deep Metric Learning | ✓ Link | 81.3 | ResNet50 + Language | 2022-03-16 |
| Calibrated neighborhood aware confidence measure for deep metric learning | | 81.2 | NED | 2020-06-08 |
| A unifying mutual information view of metric learning: cross-entropy vs. pairwise losses | ✓ Link | 81.1 | ResNet-50 + Cross-Entropy | 2020-03-19 |
| S2SD: Simultaneous Similarity-based Self-Distillation for Deep Metric Learning | ✓ Link | 81.0 | ResNet50 + S2SD | 2020-09-17 |
| Hierarchical Average Precision Training for Pertinent Image Retrieval | ✓ Link | 81.0 | HAPPIER | 2022-07-05 |
| ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis | ✓ Link | 80.7 | ResNet-50 + ProxyNCA++ | 2020-04-02 |
| Non-isotropy Regularization for Proxy-based Deep Metric Learning | ✓ Link | 80.7 | ResNet50 + NIR | 2022-03-16 |
| DAS: Densely-Anchored Sampling for Deep Metric Learning | ✓ Link | 80.59 | MS + DAS (K=8) | 2022-07-30 |
| Proxy Anchor Loss for Deep Metric Learning | ✓ Link | 80.3 | BN-Inception + Proxy-Anchor | 2020-03-31 |
| DiVA: Diverse Visual Feature Aggregation for Deep Metric Learning | ✓ Link | 79.6 | ResNet50 + DiVA | 2020-04-28 |
| Attributable Visual Similarity Learning | ✓ Link | 79.6 | ResNet50 + AVSL | 2022-03-28 |
| Towards Interpretable Deep Metric Learning with Structural Matching | ✓ Link | 79.26 | Margin + DIML | 2021-08-12 |
| Circle Loss: A Unified Perspective of Pair Similarity Optimization | ✓ Link | 78.3 | Circle Loss | 2020-02-25 |
| Improved Embeddings with Easy Positive Triplet Mining | ✓ Link | 78.3 | EPSHN(512) | 2019-04-08 |
| Cross Modal Retrieval with Querybank Normalisation | ✓ Link | 78.1 | QB-Norm+RDML | 2021-12-23 |
| MIC: Mining Interclass Characteristics for Improved Metric Learning | ✓ Link | 77.2 | ResNet50 (128) + MIC | 2019-09-25 |
| PADS: Policy-Adapted Sampling for Visual Similarity Learning | ✓ Link | 76.5 | ResNet50 (128) + PADS | 2020-03-24 |
| The Group Loss for Deep Metric Learning | ✓ Link | 75.7 | Group Loss | 2019-12-01 |