AI Paper Atlas

Research map / Benchmarks / Pitts30k

Pitts30k leaderboard

18 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR report results on Pitts30k. Best R@1: 95.8 (SAGE (8448-d), ICLR 2026).

Numbers are as each paper reports them, read by AI from its results tables and not checked by hand. Papers can differ in setting (training data, model size, extra data); the topic column shows what each paper is about. Open the paper before you compare or quote a number.

Pitts30k · R@1

18 papers, each with its best reported number for its own method; higher is better.

#R@1 ↑MethodPaperTopic
195.8SAGE (8448-d)
DINOv2
SAGE: Spatial-visual Adaptive Graph Exploration for Efficient Visual Place Recognition
ICLR 2026 · Table 1
Recognition and long-tail
295.6A2GC (33280, ViTg)
DINOv2 ViT-g/14
A2GC: Asymmetric Aggregation with Geometric Constraints for Locally Aggregated Descriptors
CVPR 2026 · Table 1
Recognition and long-tail
395SuperVLAD
DINOv2-B
SuperVLAD: Compact and Robust Image Descriptors for Visual Place Recognition
NeurIPS 2024 · Table 2
Recognition and long-tail
494.9CricaVPR
DINOv2 ViT-B/14
CricaVPR: Cross-image Correlation-aware Representation Learning for Visual Place Recognition
CVPR 2024 · Table 2
Cross-modal learning
594.5FoL-re-ranking
DINOv2 ViT-L
Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition
AAAI 2025 · Appendix Table
CNN features
694.4SelaVPR++ (performance)
DINOv2-L
SelaVPR++: Towards Seamless Adaptation of Foundation Models for Efficient Place Recognition
TPAMI 2025 · Table 2
Recognition and long-tail
794.2SemVPR (1024-dim)
DINOv2 ViT-L
Efficient Visual Place Recognition Through Multimodal Semantic Knowledge Integration
ICCV 2025 · Table 2
Multimodal LLMs
894.1ImAge
DINOv2-base-register
Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era
NeurIPS 2025 · Table 1
Attention and transformers
994EMVP-L
DINOv2 ViT-L
EMVP: Embracing Visual Foundation Model for Visual Place Recognition with Centroid-Free Probing
NeurIPS 2024 · Table 1(b)
Human and embodied modeling
1093.9EffoVPR-R
DINOv2 ViT-L/14
EffoVPR: Effective Foundation Model Utilization for Visual Place Recognition
ICLR 2025 · Comparison to state-of-the-art
CLIP and zero-shot
8 more papers on this leaderboard, the baselines reported on it and a CSV download are free with a scime account.
Sign in free to see all

Not affiliated with any conference, journal or the benchmark's authors. Corrections: contact@latenticlabs.com.