Image Retrieval On Sop
Metrics
R@1
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | R@1 |
|---|---|
| cross-batch-memory-for-embedding-learning | 80.6 |
| unicom-universal-and-compact-representation | 91.2 |
| robust-and-decomposable-average-precision-for | 86.0 |
| multi-similarity-loss-with-general-pair | 78.2 |
| attention-based-ensemble-for-deep-metric | 76.3 |
| robust-and-decomposable-average-precision-for | 83.1 |
| improved-embeddings-with-easy-positive | 78.3 |
| smooth-ap-smoothing-the-path-towards-large | 80.1 |
| rethinking-ranking-based-loss-functions-only | 81.1 |
| proxynca-revisiting-and-revitalizing-proxy | 81.4 |
| hard-aware-deeply-cascaded-embedding | 69.5 |
| combination-of-multiple-global-descriptors | 84.2 |
| making-classification-competitive-for-deep | 79.5 |
| deep-metric-learning-with-bier-boosting | 74.2 |