Image Retrieval On Par6K
評価指標
mAP
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
モデル名 | mAP | Paper Title | Repository |
|---|---|---|---|
| DELF+FT+ATT+DIR+QE | 95.7% | Large-Scale Image Retrieval with Attentive Deep Local Features | |
| R-MAC+R+QE | 86.5% | Particular object retrieval with integral max-pooling of CNN activations | |
| DIR+QE* | 93.8% | Deep Image Retrieval: Learning global representations for image search | |
| DELF+FT+ATT | 85.0% | Large-Scale Image Retrieval with Attentive Deep Local Features | |
| R-MAC | 83.0% | Particular object retrieval with integral max-pooling of CNN activations | |
| siaMAC+QE* | 85.6% | CNN Image Retrieval Learns from BoW: Unsupervised Fine-Tuning with Hard Examples | |
| Offline Diffusion | 97.8% | Efficient Image Retrieval via Decoupling Diffusion into Online and Offline Processing |
0 of 7 row(s) selected.