# ForestLPR: LiDAR Place Recognition in Forests Attentioning Multiple BEV Density Images

arXiv: https://arxiv.org/abs/2503.04475 · HTML: https://mlsanity.com/paper/2503.04475/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 62 references that are in the map.
- First posted on arXiv 6 Mar 2025.

## Builds on (most cited first, 30 of 43)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation](https://mlsanity.com/paper/1612.00593/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [Group Equivariant Convolutional Networks](https://mlsanity.com/paper/1602.07576/) (2016)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition](https://mlsanity.com/paper/1804.03492/) (2018)
- [MinkLoc3D: Point Cloud Based Large-Scale Place Recognition](https://mlsanity.com/paper/2011.04530/) (2020)
- [TransVPR: Transformer-based place recognition with multi-level attention aggregation](https://mlsanity.com/paper/2201.02001/) (2022)
- [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)
- [OverlapTransformer: An Efficient and Rotation-Invariant Transformer Network for LiDAR-Based Place Recognition](https://mlsanity.com/paper/2203.03397/) (2022)
- [PCAN: 3D Attention Map Learning Using Contextual Information for Point Cloud Based Retrieval](https://mlsanity.com/paper/1904.09793/) (2019)
- [LCDNet: Deep Loop Closure Detection and Point Cloud Registration for LiDAR SLAM](https://mlsanity.com/paper/2103.05056/) (2021)
- [SOE-Net: A Self-Attention and Orientation Encoding Network for Point Cloud based Place Recognition](https://mlsanity.com/paper/2011.12430/) (2020)
- [LoGG3D-Net: Locally Guided Global Descriptor Learning for 3D Place Recognition](https://mlsanity.com/paper/2109.08336/) (2021)
- [SegMatch: Segment based loop-closure for 3D point clouds](https://mlsanity.com/paper/1609.07720/) (2016)
- [Improving Point Cloud Based Place Recognition with Ranking-based Loss and Large Batch Training](https://mlsanity.com/paper/2203.00972/) (2022)
- [Data-efficient Large Scale Place Recognition with Graded Similarity Supervision](https://mlsanity.com/paper/2303.11739/) (2023)
- [SegMap: Segment-based mapping and localization using data-driven descriptors](https://mlsanity.com/paper/1909.12837/) (2019)
- [BEVPlace: Learning LiDAR-based Place Recognition using Bird's Eye View Images](https://mlsanity.com/paper/2302.14325/) (2023)
- [Locus: LiDAR-based Place Recognition using Spatiotemporal Higher-Order Pooling](https://mlsanity.com/paper/2011.14497/) (2020)
- [NDT-Transformer: Large-Scale 3D Point Cloud Localisation using the Normal Distribution Transform Representation](https://mlsanity.com/paper/2103.12292/) (2021)
- [Intensity Scan Context: Coding Intensity and Geometry Relations for Loop Closure Detection](https://mlsanity.com/paper/2003.05656/) (2020)
- [SVT-Net: Super Light-Weight Sparse Voxel Transformer for Large Scale Place Recognition](https://mlsanity.com/paper/2105.00149/) (2021)
- [TransLoc3D : Point Cloud based Large-scale Place Recognition using Adaptive Receptive Fields](https://mlsanity.com/paper/2105.11605/) (2021)
- [RING++: Roto-translation Invariant Gram for Global Localization on a Sparse Scan Map](https://mlsanity.com/paper/2210.05984/) (2022)

## Built on by (most cited first, 6 of 6)

- [TreeLoc: 6-DoF LiDAR Global Localization in Forests via Inter-Tree Geometric Matching](https://mlsanity.com/paper/2602.01501/) (2026)
- [Voronoi-based Second-order Descriptor with Whitened Metric in LiDAR Place Recognition](https://mlsanity.com/paper/2603.14974/) (2026)
- [Geometric-Aware Hypergraph Reasoning for Novel Class Discovery in Point Cloud Segmentation](https://mlsanity.com/paper/2606.07280/) (2026)
- [DTIF: Robust Loop Closure Detection via Delaunay Triangle Topology in Complex Forests](https://mlsanity.com/paper/2607.21138/) (2026)
- [M3GA-Wild: A Large-Scale Dataset and Benchmark for Multi-Modal Multi-session Ground-to-Aerial Place Recognition in Forests](https://mlsanity.com/paper/2609.23003/) (2026)
- [PosEviLoc: Position-Conditioned Spatial Evidence for Language-Based 3D Localization](https://mlsanity.com/paper/2609.23534/) (2026)

## Abstract (arXiv)

Place recognition is essential to maintain global consistency in large-scale localization systems. While research in urban environments has progressed significantly using LiDARs or cameras, applications in natural forest-like environments remain largely under-explored. Furthermore, forests present particular challenges due to high self-similarity and substantial variations in vegetation growth over time. In this work, we propose a robust LiDAR-based place recognition method for natural forests, ForestLPR. We hypothesize that a set of cross-sectional images of the forest's geometry at different heights contains the information needed to recognize revisiting a place. The cross-sectional images are represented by acbev density images of horizontal slices of the point cloud at different heights. Our approach utilizes a visual transformer as the shared backbone to produce sets of local descriptors and introduces a multi-BEV interaction module to attend to information at different heights adaptively. It is followed by an aggregation layer that produces a rotation-invariant place descriptor. We evaluated the efficacy of our method extensively on real-world data from public benchmarks as well as robotic datasets and compared it against the state-of-the-art (SOTA) methods. The results indicate that ForestLPR has consistently good performance on all evaluations and achieves an average increase of 7.38% and 9.11% on Recall@1 over the closest competitor on intra-sequence loop closure detection and inter-sequence re-localization, respectively, validating our hypothesis