# PROBE-X: Learning-Free Cross-FOV Place Recognition

arXiv: https://arxiv.org/abs/2607.06782 · HTML: https://mlsanity.com/paper/2607.06782/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 20 references that are in the map.
- First posted on arXiv 7 Jul 2026.

## Builds on (most cited first, 13 of 13)

- [OverlapNet: Loop Closing for LiDAR-based SLAM](https://mlsanity.com/paper/2105.11344/) (2021)
- [Scan Context++: Structural Place Recognition Robust to Rotation and Lateral Variations in Urban Environments](https://mlsanity.com/paper/2109.13494/) (2021)
- [A Survey on Global LiDAR Localization: Challenges, Advances and Open Problems](https://mlsanity.com/paper/2302.07433/) (2023)
- [LiDAR Iris for Loop-Closure Detection](https://mlsanity.com/paper/1912.03825/) (2019)
- [RING++: Roto-translation Invariant Gram for Global Localization on a Sparse Scan Map](https://mlsanity.com/paper/2210.05984/) (2022)
- [HeLiPR: Heterogeneous LiDAR Dataset for inter-LiDAR Place Recognition under Spatiotemporal Variations](https://mlsanity.com/paper/2309.14590/) (2023)
- [STD: Stable Triangle Descriptor for 3D place recognition](https://mlsanity.com/paper/2209.12435/) (2022)
- [GrandTour: A Legged Robotics Dataset in the Wild for Multi-Modal Perception and State Estimation](https://mlsanity.com/paper/2602.18164/) (2026)
- [Narrowing your FOV with SOLiD: Spatially Organized and Lightweight Global Descriptor for FOV-constrained LiDAR Place Recognition](https://mlsanity.com/paper/2408.07330/) (2024)
- [HeLiOS: Heterogeneous LiDAR Place Recognition via Overlap-based Learning and Local Spherical Transformer](https://mlsanity.com/paper/2501.18943/) (2025)
- [UniLGL: Learning Uniform Place Recognition for FOV-limited/Panoramic LiDAR Global Localization](https://mlsanity.com/paper/2507.12194/) (2025)
- [Efficiently Closing Loops in LiDAR-Based SLAM Using Point Cloud Density Maps](https://mlsanity.com/paper/2501.07399/) (2025)
- [PROBE: Probabilistic Occupancy BEV Encoding with Analytical Translation Robustness for 3D Place Recognition](https://mlsanity.com/paper/2603.05965/) (2026)

## Abstract (arXiv)

Under field-of-view (FOV) mismatch, pooling LiDAR features over unequal angular support can distort compact retrieval keys and exclude correct matches before geometric verification. We present PROBE-X, a learning-free method for single-scan cross-FOV place recognition. Building on PROBE's probabilistic occupancy representation, which models translation uncertainty, PROBE-X constructs ring-mean retrieval keys using angular masks conditioned on up to four fixed hypotheses for the unknown relative heading. Candidate lists are merged by maximum cosine similarity. Each retained candidate is aligned under the hypotheses that retrieved it and assigned a geometric score over the angular overlap at the refined heading. Across controlled-FOV and heterogeneous-LiDAR experiments, PROBE-X achieves higher Recall@1 and PR-AUC than the evaluated single-scan baselines in most settings with limited-FOV queries and panoramic database scans. On the controlled asymmetric-FOV sweep, single-scan PROBE-X also outperforms the baselines evaluated with up to ten accumulated scans on both metrics. In two ablation settings, removing the angular masks from retrieval-key construction reduces candidate coverage and Recall@1, even when alignment and scoring still account for angular support. Project page: https://sites.google.com/view/probex-pr