| Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era | ✓ Link | 97.7 | 99.3 | 99.6 | ImAge (6144D, unified dataset) | 2025-11-08 |
| SelaVPR++: Towards Seamless Adaptation of Foundation Models for Efficient Place Recognition | ✓ Link | 97.6 | 99.2 | | SelaVPR++ (DINOv2-B, BoQ12288D, unified training) | 2025-02-23 |
| Query-Based Adaptive Aggregation for Multi-Dataset Joint Training Toward Universal Visual Place Recognition | | 96.7 | | | QAA-DINOv2-B-8192 | 2025-07-04 |
| AdaptVPR: Route-Aware Hard Positive Generation for Robust Visual Place Recognition | ✓ Link | 95.9 | 98.5 | 99.1 | BoQ + AdaptVPR | 2026-09-03 |
| EffoVPR: Effective Foundation Model Utilization for Visual Place Recognition | | 95.0 | 98.6 | | EffoVPR | 2024-05-28 |
| SAGE: Spatial-visual Adaptive Graph Exploration for Efficient Visual Place Recognition | ✓ Link | 94.8 | 98.2 | 98.9 | SAGE ViT-L (without cross-image encoder) | 2025-09-30 |
| Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition | ✓ Link | 92.6 | 96.9 | 98.0 | FoL | 2025-04-14 |
| SuperPlace: The Renaissance of Classical Feature Aggregation for Visual Place Recognition in the Era of Foundation Models | ✓ Link | 91.4 | | | SuperPlace SP-NVL-FT2 (DINOv2-B, GPMS, 8192D, 322px) | 2025-06-16 |
| BoQ: A Place is Worth a Bag of Learnable Queries | ✓ Link | 90.6 | 96.0 | 97.5 | BoQ | 2024-05-12 |
| EDTformer: An Efficient Decoder Transformer for Visual Place Recognition | ✓ Link | 88.3 | 95.3 | 97 | EDTformer (4096D, 2decoder blocks, 64learned tokens, LoPA rank4) | 2024-12-01 |
| Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition | ✓ Link | 87.8 | 94.5 | 96.4 | FoL-global | 2025-04-14 |
| Towards Seamless Adaptation of Pre-trained Models for Visual Place Recognition | ✓ Link | 86.6 | 94.0 | | SelaVPR | 2024-02-22 |
| Optimal Transport Aggregation for Visual Place Recognition | ✓ Link | 85.2 | 98.5 | | DINOv2 SALAD (1-frame thr.) | 2023-11-27 |
| BoQ: A Place is Worth a Bag of Learnable Queries | ✓ Link | 83.1 | | | BoQ (ResNet-50) | 2024-05-12 |
| D²-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation | ✓ Link | 78 | 89.1 | | D2-VPR (no encoder) | 2025-11-16 |
| MixVPR: Feature Mixing for Visual Place Recognition | ✓ Link | 76.0 | 89.2 | | MixVPR | 2023-03-03 |
| EigenPlaces: Training Viewpoint Robust Models for Visual Place Recognition | ✓ Link | 71.2 | 83.8 | 88.1 | EigenPlaces (ResNet-50, 2048D) | 2023-08-21 |
| Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition | ✓ Link | 58.4 | 74.6 | | Patch-NetVLAD | 2021-03-02 |
| Global Proxy-based Hard Mining for Visual Place Recognition | ✓ Link | 44.9 | 50.2 | | NetVLAD (with GPM) | 2023-02-28 |
| GSV-Cities: Toward Appropriate Supervised Visual Place Recognition | ✓ Link | 38.5 | 53.9 | | Conv-AP | 2022-10-19 |
| A Hierarchical Dual Model of Environment- and Place-Specific Utility for Visual Place Recognition | ✓ Link | 38.2 | 54.8 | | HEAPUtil | 2021-07-06 |
| SelaVPR++: Towards Seamless Adaptation of Foundation Models for Efficient Place Recognition | ✓ Link | | | 99.5 | SelaVPR++ (DINOv2-L, float4096D direct retrieval) | 2025-02-23 |
| D²-VPR: A Parameter-efficient Visual-foundation-model-based Visual Place Recognition Method via Knowledge Distillation and Deformable Aggregation | ✓ Link | | | 91.9 | D2-VPR (released no encoder) | 2025-11-16 |