Research map / Label-efficient and robust learning / Recognition and long-tail
Lost in the Tail: Addressing Geographic Imbalance in Urban Visual Place Recognition
Zoey Shu, Jiacheng Yang, Yang Lu, Waishan Qiu, Chuan Li, Da Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace cross-entropy with LB loss (L2 retrieval) (ablation) | DINOv2 mixed pipeline · SF-XL test v1 | Recall@1 (%): 75.8 -> 88.0 | Improved |
| replace L2 with CFD retrieval (ablation) | DINOv2 + LB loss · SF-XL test v1 | Recall@1 (%): 88.0 -> 89.7 | Improved |
| replace L2 with CFD retrieval (ablation) | DINOv2 + LB loss · SF-XL test v2 | Recall@1 (%): 93.8 -> 94.3 | Improved |
| add LB loss to BoQ | BoQ · SF-XL test v1 | Recall@1 (%): 83.7 -> 88.8 | Improved |
| add LB loss to SALAD | SALAD · SF-XL test v1 | Recall@1 (%): 87.6 -> 88.0 | Improved |
| add LB loss to CosPlace | CosPlace · SF-XL test v1 | Recall@1 (%): 74.4 -> 76.1 | Improved |
| add LB loss to CosPlace | CosPlace · SF-XL test v2 | Recall@1 (%): 90.0 -> 90.0 | No clear effect |
| add LB loss to D&C | D&C · SF-XL test v1 | Recall@1 (%): 44.1 -> 48.2 | Improved |
| add LB loss to D&C mixed pipeline | D&C (mixed) · SF-XL test v1 | Recall@1 (%): 55.1 -> 57.4 | Improved |
| add LB loss to BoQ | BoQ · MSLS | Recall@1 (%): 91.2 -> 93.7 | Improved |
| add LB loss to BoQ | BoQ · Pitts30k | Recall@1 (%): 92.6 -> 92.9 | Improved |
| add LB loss to BoQ | BoQ · Nordland | Recall@1 (%): 81.3 -> 83.7 | Improved |
Closest papers by meaning
- Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition (AAAI 2025 · similarity 0.90)
- Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era (NeurIPS 2025 · similarity 0.89)
- Optimal Transport Aggregation for Visual Place Recognition (CVPR 2024 · similarity 0.89)
- MutualVPR: A Mutual Learning Framework for Resolving Supervision Inconsistencies via Adaptive Clustering (NeurIPS 2025 · similarity 0.88)
- EffoVPR: Effective Foundation Model Utilization for Visual Place Recognition (ICLR 2025 · similarity 0.88)
- On the Estimation of Image-matching Uncertainty in Visual Place Recognition (CVPR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
