| Paper | Code | Recall@1 | Recall@10 | ModelName | ReleaseDate |
|---|---|---|---|---|---|
| SAGE: Spatial-visual Adaptive Graph Exploration for Efficient Visual Place Recognition | ✓ Link | 0.599 | 0.779 | SAGE ViT-L (without cross-image encoder) | 2025-09-30 |
| AdaptVPR: Route-Aware Hard Positive Generation for Robust Visual Place Recognition | ✓ Link | 0.562 | 0.719 | EDTformer + AdaptVPR | 2026-09-03 |
| UniPR-3D: Towards Universal Visual Place Recognition with Visual Geometry Grounded Transformer | ✓ Link | 0.548 | 0.745 | UniPR-3D* (MegaLoc training setting, single frame) | 2025-12-24 |
| MegaLoc: One Retrieval to Place Them All | ✓ Link | 0.528 | 0.738 | MegaLoc (8448D) | 2025-02-24 |
| SuperPlace: The Renaissance of Classical Feature Aggregation for Visual Place Recognition in the Era of Foundation Models | ✓ Link | 0.453 | SuperPlace SP-NVL-FT2 (DINOv2-B, GPMS, 8192D, 322px) | 2025-06-16 |