Image Retrieval On Google Landmarks Dataset
평가 지표
mAP@100
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | mAP@100 | Paper Title | Repository |
|---|---|---|---|
| ResNet101+ArcFace GLDv2-train-clean | 24.15 | Google Landmarks Dataset v2 -- A Large-Scale Benchmark for Instance-Level Recognition and Retrieval | |
| UNICOM-ViT-B-16-512px | 35.7 | Unicom: Universal and Compact Representation Learning for Image Retrieval | |
| AMES | 37.8 | AMES: Asymmetric and Memory-Efficient Similarity Estimation for Instance-level Retrieval | |
| UNICOM-ViT-L-14-512px | 36.4 | Unicom: Universal and Compact Representation Learning for Image Retrieval |
0 of 4 row(s) selected.