Image Retrieval on RPar+R1M Medium
87.2mAPSG + LOCORE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SG + LOCOREGlobal=SG, Local=DINOv2, Re-rank=SG + LOCORE2025.03 | 87.2 | — | |
| LOCOREGlobal=SG, Local=DINOv2, Re-rank=LOCORE2025.03 | 86.8 | — | |
| SG + LOCOREGlobal=SG, Local=CVNet, Re-rank=SG + LOCORE2025.03 | 86.2 | — | |
| AMESGlobal=SG, Local=DINOv2, Re-rank=AMES, Hidden Size=7682025.03 | 85.8 | — | |
| RRTGlobal=SG, Local=DINOv2, Re-rank=RRT, Hidden Size=7682025.03 | 85.5 | — | |
| SG-RerankGlobal=SG, Local=N/A, Re-rank=SG-Rerank2025.03 | 85.2 | — | |
| SG-RerankGlobal=SG, Local=N/A, Re-rank=SG-Rerank2025.03 | 84.9 | — | |
| R2FormerGlobal=SG, Local=DINOv2, Re-rank=R2Former, Hidden Size=7682025.03 | 84.9 | — | |
| AMESGlobal=SG, Local=CVNet, Re-rank=AMES2025.03 | 84.6 | — | |
| LOCOREGlobal=SG, Local=CVNet, Re-rank=LOCORE2025.03 | 84.6 | — | |
| Baseline (SuperGlobal)Global=SG, Local=N/A, Re-rank=N/A2025.03 | 83.9 | — | |
| R2FormerGlobal=SG, Local=CVNet, Re-rank=R2Former2025.03 | 83.8 | — | |
| BaselineGlobal=SG, Local=N/A, Re-rank=N/A2025.03 | 83.6 | — | |
| RRTGlobal=SG, Local=CVNet, Re-rank=RRT2025.03 | 83.6 | — | |
| SG + LOCOREGlobal=DINOv2, Local=DINOv2, Re-rank=SG + LOCORE2025.03 | 81.9 | — | |
| LOCOREGlobal=DINOv2, Local=DINOv2, Re-rank=LOCORE2025.03 | 80.9 | — | |
| AMESGlobal=DINOv2, Local=DINOv2, Re-rank=AMES, Hidden Size=7682025.03 | 79.9 | — | |
| SG-RerankGlobal=DINOv2, Local=N/A, Re-rank=SG-Rerank2025.03 | 79.8 | — | |
| RRTGlobal=DINOv2, Local=DINOv2, Re-rank=RRT, Hidden Size=7682025.03 | 79.6 | — | |
| R2FormerGlobal=DINOv2, Local=DINOv2, Re-rank=R2Former, Hidden Size=7682025.03 | 78.6 | — | |
| BaselineGlobal=DINOv2, Local=N/A, Re-rank=N/A2025.03 | 77 | — | |
| CiDeR-FTTraining Set=RGLDv2-clean, Fine-tuned=true2024.04 | 61.6 | 94.3 | |
| CiDeRTraining Set=RGLDv2-clean, Fine-tuned=false2024.04 | 59 | 95.1 | |
| GLAMTraining Set=RGLDv2-clean, Fine-tuned=false2024.04 | 58.6 | 91.7 | |
| SOLARTraining Set=RGLDv2-clean, Fine-tuned=true2024.04 | 57.6 | 92 | |
| DOLGTraining Set=RGLDv2-clean, Fine-tuned=false2024.04 | 57.3 | 92 | |
| CiDeR-FTTraining Set=SfM-120k, Fine-tuned=true2024.04 | 56.9 | 95.9 | |
| Yokoo et al.Training Set=RGLDv2-clean, Fine-tuned=false2024.04 | 54.2 | 90.3 | |
| Filip et al.Training Set=SfM-120k, Fine-tuned=true2024.04 | 52.3 | 95.3 | |
| CiDeRTraining Set=SfM-120k, Fine-tuned=false2024.04 | 51.4 | 95.7 | |
| TokenTraining Set=RGLDv2-clean, Fine-tuned=false2024.04 | 44.1 | 86.9 |