pith. sign in

arxiv: 2501.07399 · v2 · submitted 2025-01-13 · 💻 cs.RO

Efficiently Closing Loops in LiDAR-Based SLAM Using Point Cloud Density Maps

Saurabh Gupta , Tiziano Guadagnino , Benedikt Mersch , Niklas Trekel , Meher V. R. Malladi , Cyrill Stachniss This is my paper

Pith reviewed 2026-05-23 05:43 UTC · model grok-4.3

classification 💻 cs.RO
keywords loop closure detectionLiDAR SLAMplace recognitionbird's-eye-view projectionORB featurespoint cloud densityground alignmentperceptual aliasing
0
0 comments X

The pith

A loop closure pipeline aligns LiDAR maps to ground then matches ORB features on density-preserving bird's-eye views to detect places across different sensors.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces a loop closure detection method that turns LiDAR scans into local maps, aligns those maps to a ground plane to accommodate both flat and uneven robot motion, and converts the aligned maps into bird's-eye projections that keep the original point density. ORB descriptors are extracted from the projections, stored in a binary search tree for fast lookup, and filtered by a self-similarity check to reduce false matches in repetitive scenes. A sympathetic reader would care because the pipeline is presented as working without change on LiDARs that differ in scan pattern, field of view, and resolution, thereby supporting drift correction and multi-robot map alignment in outdoor settings.

Core claim

Generating local maps from raw LiDAR scans, applying a ground alignment step that works for planar and non-planar trajectories, forming density-preserving bird's-eye-view images, extracting ORB features, indexing them in a binary search tree, and pruning self-similar entries produces reliable loop closures that remain accurate across sensor types and motion profiles, as shown by experiments on public and self-recorded datasets for localization and cross-platform map merging.

What carries the argument

Ground alignment module followed by density-preserving bird's-eye-view projection that feeds ORB feature extraction and binary-search-tree retrieval with self-similarity pruning.

If this is right

  • Loop closures are detected accurately on both public benchmarks and self-recorded outdoor sequences.
  • Long-term localization remains consistent because drift is corrected without sensor-specific tuning.
  • Multiple maps recorded by different platforms can be aligned into a single consistent map.
  • The method stays effective under both planar and non-planar motion and in environments with repetitive structure.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Robots carrying dissimilar LiDARs could exchange and merge maps without extra calibration steps.
  • The same projection and pruning steps might be tested on indoor sequences where vertical structure is richer.
  • Replacing ORB with learned descriptors could be compared directly on the same density-preserving views.

Load-bearing premise

Ground alignment plus density-preserving projections keep enough unique geometric detail that ORB features can still match the same place when sensor resolution, field of view, and motion differ substantially.

What would settle it

Running the pipeline on a dataset that pairs two LiDARs with visibly different resolutions or fields of view and counting the fraction of incorrect or missed loop closures would directly test whether the claimed sensor-agnostic performance holds.

Figures

Figures reproduced from arXiv: 2501.07399 by Benedikt Mersch, Cyrill Stachniss, Meher V. R. Malladi, Niklas Trekel, Saurabh Gupta, Tiziano Guadagnino.

Figure 1
Figure 1. Figure 1: An example of loop closures detected between two sequences recorded with different LiDAR sensor platforms with a revisit interval [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: An overview of our pipeline for loop closure detection and alignment. Given an input stream of point clouds [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: A block diagram showcasing the composition of a local [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: A comparison of data from LiDAR sensors with different scanning patterns and field of views. The first row shows a single scan [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Effect of ground alignment on a local map generated from a handheld LiDAR sensor with non-planar motion. The colors of the points i [PITH_FULL_IMAGE:figures/full_fig_p006_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: A BEV density image of a local map with darker pixels [PITH_FULL_IMAGE:figures/full_fig_p007_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: An example of self-similarity feature pruning on the ORB [PITH_FULL_IMAGE:figures/full_fig_p007_7.png] view at source ↗
Figure 9
Figure 9. Figure 9: Two local maps (in red and blue) detected as loop closure [PITH_FULL_IMAGE:figures/full_fig_p008_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: An illustration of the conversion from loop closures between [PITH_FULL_IMAGE:figures/full_fig_p010_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: The precision-recall curves of state-of-the-art baselines and our approach for single-session loop closure detection. [PITH_FULL_IMAGE:figures/full_fig_p012_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: Effect of the ground alignment strategy on the quality of alignment [PITH_FULL_IMAGE:figures/full_fig_p015_12.png] view at source ↗
Figure 13
Figure 13. Figure 13: Perceptual aliasing from the HeLiPR Bridge sequence. [PITH_FULL_IMAGE:figures/full_fig_p016_13.png] view at source ↗
Figure 14
Figure 14. Figure 14: We use our loop closure pipeline to align the three KAIST sequences respectively to the Riverside03 sequence from MulRan dataset. [PITH_FULL_IMAGE:figures/full_fig_p017_14.png] view at source ↗
read the original abstract

Consistent maps are key for most autonomous mobile robots, and they often use SLAM approaches to build such maps. Loop closures via place recognition help to maintain accurate pose estimates by mitigating global drift, and are thus key for realizing an effective SLAM system. This paper presents a robust loop closure detection pipeline for outdoor SLAM with LiDAR-equipped robots. Our method handles various LiDAR sensors with different scanning patterns, fields of view, and resolutions. It generates local maps from LiDAR scans and aligns them using a ground alignment module to handle both planar and non-planar motion of the LiDAR, ensuring applicability across platforms. The method uses density-preserving bird's-eye-view projections of these local maps and extracts ORB feature descriptors for place recognition. It stores the feature descriptors in a binary search tree for efficient retrieval, and self-similarity pruning addresses perceptual aliasing in repetitive environments. Extensive experiments on public and self-recorded datasets demonstrate accurate loop closure detection, long-term localization, and cross-platform multi-map alignment, agnostic to the LiDAR scanning patterns, fields of view, and motion profiles. We provide the code for our pipeline as open-source software at https://github.com/PRBonn/MapClosures.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper claims to introduce a robust loop closure detection pipeline for outdoor LiDAR SLAM that handles sensors with varying scan patterns, FOVs, and resolutions. It generates local maps, applies a ground alignment module for planar/non-planar motion, creates density-preserving BEV projections, extracts ORB descriptors stored in a binary search tree, and uses self-similarity pruning against perceptual aliasing. Experiments on public and self-recorded datasets are reported to demonstrate accurate loop closure, long-term localization, and cross-platform multi-map alignment, with open-source code released.

Significance. If the cross-sensor claims hold, the work supplies a practical, efficient place-recognition module for LiDAR SLAM that could support multi-platform mapping and drift correction in outdoor robotics. The open-source release is a clear strength that enables direct reproducibility and extension.

major comments (2)
  1. [Method (ground alignment and BEV projection)] The central generalization claim—that ground alignment plus density-preserving BEV projection preserves sufficient place-specific geometry for reliable ORB matching across differing LiDAR scan patterns, FOVs, and resolutions—is load-bearing. The projection step necessarily removes vertical relief and cannot synthesize missing rays from narrower-FOV or lower-resolution sensors, yet the manuscript provides no ablation quantifying the resulting loss in feature distinctiveness or matching precision under these conditions.
  2. [Experiments] While the abstract asserts that extensive experiments on public and self-recorded datasets support the claims, the absence of reported quantitative metrics (e.g., precision-recall, cross-sensor retrieval rates), ablation studies isolating the BEV step, or explicit error analysis on sensor-variation cases leaves the strength of evidence for the invariance claim unverifiable.
minor comments (1)
  1. [Abstract] The abstract would benefit from naming the specific public datasets used, to give readers immediate context for the reported results.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their thorough review and constructive comments on our manuscript. We address each of the major comments below and are prepared to make revisions to strengthen the presentation of our results.

read point-by-point responses

  1. Referee: [Method (ground alignment and BEV projection)] The central generalization claim—that ground alignment plus density-preserving BEV projection preserves sufficient place-specific geometry for reliable ORB matching across differing LiDAR scan patterns, FOVs, and resolutions—is load-bearing. The projection step necessarily removes vertical relief and cannot synthesize missing rays from narrower-FOV or lower-resolution sensors, yet the manuscript provides no ablation quantifying the resulting loss in feature distinctiveness or matching precision under these conditions.

    Authors: We acknowledge that the manuscript would benefit from an explicit ablation study to quantify the impact of the ground alignment and density-preserving BEV projection on feature matching across different sensor configurations. The current evaluation demonstrates the method's performance through successful loop closure detection and cross-platform alignment on datasets collected with varying LiDAR sensors. In the revised version, we will include an ablation analysis to better support the generalization claims. revision: yes

  2. Referee: [Experiments] While the abstract asserts that extensive experiments on public and self-recorded datasets support the claims, the absence of reported quantitative metrics (e.g., precision-recall, cross-sensor retrieval rates), ablation studies isolating the BEV step, or explicit error analysis on sensor-variation cases leaves the strength of evidence for the invariance claim unverifiable.

    Authors: We agree that providing precision-recall curves, cross-sensor retrieval rates, and ablations would make the experimental evidence more robust and verifiable. The manuscript includes quantitative results on loop closure accuracy and qualitative demonstrations of long-term localization and multi-map alignment. We will expand the experiments section with the suggested metrics and analyses in the revision. revision: yes

Circularity Check

0 steps flagged

No circularity; pipeline composes standard independent primitives

full rationale

The paper presents an engineering pipeline that generates local maps, applies ground alignment, performs density-preserving BEV projection, extracts ORB descriptors, stores them in a BST, and applies self-similarity pruning. None of these steps are defined in terms of the claimed performance metrics, nor are any parameters fitted on the evaluation data and then re-used as 'predictions.' No self-citations appear as load-bearing uniqueness results, and the abstract and method description contain no equations that reduce the output to the input by construction. The approach is therefore self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Review performed on abstract only; no explicit free parameters, axioms, or invented entities are identifiable from the given text. The approach uses standard, previously published components.

pith-pipeline@v0.9.0 · 5768 in / 1206 out tokens · 47214 ms · 2026-05-23T05:43:04.532918+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

71 extracted references · 71 canonical work pages · 1 internal anchor

  1. [1]

    Arandjelovic, P

    R. Arandjelovic, P. Gronat, A. Torii, T. Pajdla, and J. Sivic. NetVLAD: CNN Architecture for Weakly Supervised Place Recognition. In Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , 2016

    work page 2016

  2. [2]

    Besl and N

    P. Besl and N. McKay. A Method for Registration of 3D Shapes. IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI) , 14(2):239–256, 1992

    work page 1992

  3. [3]

    Bosse, P

    M. Bosse, P. Newman, J. Leonard, and S. Teller. Simultaneous Localization and Map Building in Large-Scale Cyclic Environments Using the Atlas Framework. Intl. Journal of Robotics Research (IJRR) , 23(12):1113 – 1139, 2004

    work page 2004

  4. [4]

    Bosse and R

    M. Bosse and R. Zlot. Place Recognition Using Keypoint V oting in Large 3D Lidar Datasets. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2013

    work page 2013

  5. [5]

    Carlevaris-Bianco, A

    N. Carlevaris-Bianco, A. Ushani, and R. Eustice. University of Michigan North Campus Long-term Vision and LiDAR Dataset. Intl. Journal of Robotics Research (IJRR) , 35(9):1023–1035, 2016

    work page 2016

  6. [6]

    X. Chen, T. L ¨abe, A. Milioto, T. R ¨ohling, O. Vysotska, A. Haag, J. Behley, and C. Stachniss. OverlapNet: Loop Closing for LiDAR- based SLAM. In Proc. of Robotics: Science and Systems (RSS) , 2020

    work page 2020

  7. [7]

    Chen and G

    Y . Chen and G. Medioni. Object Modelling by Registration of Multiple Range Images. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 1991

    work page 1991

  8. [8]

    Cummins and P

    M. Cummins and P. Newman. FAB-MAP: Probabilistic Localization and Mapping in the Space of Appearance. Intl. Journal of Robotics Research (IJRR), 27(6):647–665, 2008

    work page 2008

  9. [9]

    Dellenbach, J

    P. Dellenbach, J. Deschaud, B. Jacquet, and F. Goulette. CT-ICP Real- Time Elastic LiDAR Odometry with Loop Closure. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2022

    work page 2022

  10. [10]

    Dub ´e, A

    R. Dub ´e, A. Cramariuc, D. Dugas, J. Nieto, R. Siegwart, and C. Cadena. SegMap: 3D Segment Mapping using Data-Driven Descriptors. In Proc. of Robotics: Science and Systems (RSS) , 2018

    work page 2018

  11. [11]

    Ferrari, L.D

    S. Ferrari, L.D. Giammarino, L. Brizi, and G. Grisetti. MAD-ICP: It Is All About Matching Data–Robust and Informed LiDAR Odometry. IEEE Robotics and Automation Letters (RA-L) , 9(11):9175–9182, 2024

    work page 2024

  12. [12]

    Fontana, G

    S. Fontana, G. Agamennoni, R. Siegwart, and D. Sorrenti. Point Clouds Registration with Probabilistic Data Association. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2016

    work page 2016

  13. [13]

    Frome, D

    A. Frome, D. Huber, R. Kolluri, T. B ¨ulow, and J. Malik. Recognizing Objects in Range Data Using Regional Point Descriptors. In Proc. of the Europ. Conf. on Computer Vision (ECCV) , 2004

    work page 2004

  14. [14]

    Galvez-Lopez and J.D

    D. Galvez-Lopez and J.D. Tardos. Bags of Binary Words for Fast Place Recognition in Image Sequences. IEEE Trans. on Robotics (TRO) , 28(5):1188–1197, 2012

    work page 2012

  15. [15]

    Guadagnino, X

    T. Guadagnino, X. Chen, M. Sodano, J. Behley, G. Grisetti, and C. Stachniss. Fast Sparse LiDAR Odometry Using Self-Supervised Feature Selection on Intensity Images. IEEE Robotics and Automation Letters (RA-L) , 7(3):7597–7604, 2022

    work page 2022

  16. [16]

    Gupta, T

    S. Gupta, T. Guadagnino, B. Mersch, I. Vizzo, and C. Stachniss. Effectively Detecting Loop Closures using Point Cloud Density Maps. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2024

    work page 2024

  17. [17]

    L. He, X. Wang, and H. Zhang. M2DP: A Novel 3D Point Cloud Descriptor and Its Application in Loop Closure Detection. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2016

    work page 2016

  18. [18]

    Jiang and S

    B. Jiang and S. Shen. Contour Context Abstract Structural Distribution for 3D LiDAR Loop Detection and Metric Pose Estimation. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2023

    work page 2023

  19. [19]

    Johnson and M

    A. Johnson and M. Hebert. Using Spin Images for Effcient Object Recognition in Cluttered 3D Scenes. IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI) , 21(5):433–449, 1999

    work page 1999

  20. [20]

    M. Jung, W. Yang, D. Lee, H. Gil, G. Kim, and A. Kim. HeLiPR: Heterogeneous LiDAR Dataset for inter-LiDAR Place Recognition under Spatiotemporal Variations. Intl. Journal of Robotics Research (IJRR) , 43(12):1867–1883, 2024

    work page 2024

  21. [21]

    W. Kabsch. A Solution for the Best Rotation to Relate Two Sets of Vectors. Acta Crystallographica Section A: Crystal Physics, Diffraction, Theoretical and General Crystallography , 32(5):922–923, 1976

    work page 1976

  22. [22]

    G. Kim, S. Choi, and A. Kim. Scan Context++: Structural Place Recog- nition Robust to Rotation and Lateral Variations in Urban Environments. IEEE Trans. on Robotics (TRO) , 38(2):21–27, 2021

    work page 2021

  23. [23]

    G. Kim, B. Park, and A. Kim. 1-day learning, 1-year localization: Long- term LiDAR Localization Using Scan Context Image. IEEE Robotics and Automation Letters (RA-L) , 4(2):1948–1955, 2019

    work page 1948

  24. [24]

    G. Kim, Y . Park, Y . Cho, J. Jeong, and A. Kim. Mulran: Multimodal Range Dataset for Urban Place Recognition. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2020. 19

    work page 2020

  25. [25]

    Kim and A

    G. Kim and A. Kim. Scan Context: Egocentric Spatial Descriptor for Place Recognition within 3D Point Cloud Map. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2018

    work page 2018

  26. [26]

    H. Kim, J. Choi, T. Sim, G. Kim, and Y . Cho. Narrowing Your FOV With SOLiD: Spatially Organized and Lightweight Global Descriptor for FOV-Constrained LiDAR Place Recognition. IEEE Robotics and Automation Letters (RA-L) , 9(11):9645–9652, 2024

    work page 2024

  27. [27]

    Komorowski

    J. Komorowski. MinkLoc3D: Point Cloud Based Large-Scale Place Recognition. In Proc. of the IEEE Winter Conf. on Applications of Computer Vision (WACV) , 2021

    work page 2021

  28. [28]

    K ¨ummerle, G

    R. K ¨ummerle, G. Grisetti, H. Strasdat, K. Konolige, and W. Burgard. g2o: A General Framework for Graph Optimization. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2011

    work page 2011

  29. [29]

    S. Lee, H. Lim, and H. Myung. Patchwork++: Fast and Robust Ground Segmentation Solving Partial Under-segmentation Using 3D Point Cloud. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2022

    work page 2022

  30. [30]

    Li and H

    Y . Li and H. Li. LiDAR-Based Initial Global Localization Using Two- Dimensional (2D) Submap Projection Image (SPI). In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2021

    work page 2021

  31. [31]

    H. Lim, O. Minho, and H. Myung. Patchwork: Concentric Zone-based Region-wise Ground Segmentation with Ground Likelihood Estimation Using a 3D LiDAR Sensor. IEEE Robotics and Automation Letters (RA-L), 6(4):6458–6465, 2021

    work page 2021

  32. [32]

    D. Lowe. Distinctive Image Features from Scale-Invariant Keypoints. Intl. Journal of Computer Vision (IJCV) , 60(2):91–110, 2004

    work page 2004

  33. [33]

    S. Lu, X. Xu, H. Yin, Z. Chen, R. Xiong, and Y . Wang. One RING to Rule Them All: Radon Sinogram for Place Recognition, Orientation and Translation Estimation. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2022

    work page 2022

  34. [34]

    L. Luo, S. Cao, B. Han, H. Shen, and J. Li. BVMatch: Lidar-Based Place Recognition Using Bird’s-Eye View Images. IEEE Robotics and Automation Letters (RA-L) , 6(3):6076–6083, 2021

    work page 2021

  35. [35]

    L. Luo, S. Cao, Z. Sheng, and H. Shen. LiDAR-Based Global Localization Using Histogram of Orientations of Principal Normals. IEEE Trans. on Intelligent V ehicles (TIV) , 7(3):771–782, 2022

    work page 2022

  36. [36]

    L. Luo, S. Zheng, Y . Li, Y . Fan, B. Yu, S. Cao, J. Li, and H. Shen. BEVPlace: Learning LiDAR-based Place Recognition using Bird’s Eye View Images. In Proc. of the IEEE/CVF Intl. Conf. on Computer Vision (ICCV), 2023

    work page 2023

  37. [37]

    J. Ma, J. Zhang, J. Xu, R. Ai, W. Gu, and X. Chen. OverlapTransformer: An Efficient and Yaw-Angle-Invariant Transformer Network for LiDAR- Based Place Recognition. IEEE Robotics and Automation Letters (RA- L), 7(3):6958–6965, 2022

    work page 2022

  38. [38]

    Magnusson, H

    M. Magnusson, H. Andreasson, A. Nuechter, Achim, and J. Lilienthal. Appearance-Based Loop Detection from 3D Laser Data Using the Normal Distributions Transform. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2009

    work page 2009

  39. [39]

    Mendes, P

    E. Mendes, P. Koch, and S. Lacroix. ICP-based Pose-Graph SLAM. In Proc. of the IEEE Intl. Symp. on Safety, Security, and Rescue Robotics (SSRR), 2016

    work page 2016

  40. [40]

    Mur-Artal, J

    R. Mur-Artal, J. Montiel, and J.D. Tardos. ORB-SLAM: A Versatile and Accurate Monocular SLAM System. IEEE Trans. on Robotics (TRO) , 31(5):1147–1163, 2015

    work page 2015

  41. [41]

    Paigwar, ¨O

    A. Paigwar, ¨O. Erkent, D. Sierra-Gonzalez, and C. Laugier. GndNet: Fast Ground Plane Estimation and Point Cloud Segmentation for Au- tonomous Vehicles. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2020

    work page 2020

  42. [42]

    C.R. Qi, H. Su, K. Mo, and L.J. Guibas. PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation. In Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , 2017

    work page 2017

  43. [43]

    C. Qi, K. Yi, H. Su, and L.J. Guibas. PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space. In Proc. of the Conf. on Neural Information Processing Systems (NeurIPS) , 2017

    work page 2017

  44. [44]

    R ¨ohling, J

    T. R ¨ohling, J. Mack, and D. Schulz. A Fast Histogram-Based Similarity Measure for Detecting Loop Closures in 3-D LIDAR Data. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2015

    work page 2015

  45. [45]

    Rottmann, R

    N. Rottmann, R. Bruder, A. Schweikard, and E. Rueckert. Loop Closure Detection in Closed Environments. In Proc. of the Europ. Conf. on Mobile Robotics (ECMR) , 2019

    work page 2019

  46. [46]

    Rublee, V

    E. Rublee, V . Rabaud, K. Konolige, and G. Bradski. ORB: An Efficient Alternative to SIFT or SURF. In Proc. of the IEEE Intl. Conf. on Computer Vision (ICCV) , 2011

    work page 2011

  47. [47]

    R. Rusu, N. Blodow, Z. Marton, and M. Beetz. Aligning Point Cloud Views using Persistent Feature Histograms. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 09 2008

    work page 2008

  48. [48]

    R. Rusu, N. Blodow, and M. Beetz. Fast Point Feature Histograms (FPFH) for 3D Registration. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2009

    work page 2009

  49. [49]

    Salti, F

    S. Salti, F. Tombari, and L. Stefano. SHOT: Unique Signatures of Histograms for Surface and Texture Description. Journal of Computer Vision and Image Understanding (CVIU) , 125:251–264, 2014

    work page 2014

  50. [50]

    Schlegel and G

    D. Schlegel and G. Grisetti. HBST: A Hamming Distance Embedding Binary Search Tree for Visual Place Recognition. IEEE Robotics and Automation Letters (RA-L) , 3:3741–3748, 2018

    work page 2018

  51. [51]

    T. Shan, B. Englot, F. Duarte, C. Ratti, and D. Rus. Robust Place Recognition using an Imaging Lidar. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2021

    work page 2021

  52. [52]

    T. Shan, B. Englot, D. Meyers, W. Wang, C. Ratti, and D. Rus. LIO-SAM: Tightly-coupled Lidar Inertial Odometry via Smoothing and Mapping. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS), 2020

    work page 2020

  53. [53]

    Steder, G

    B. Steder, G. Grisetti, and W. Burgard. Robust Place Recognition for 3D Range Data Based on Point Features. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2010

    work page 2010

  54. [54]

    Steder, M

    B. Steder, M. Ruhnke, S. Grzonka, and W. Burgard. Place Recognition in 3D Scans Using a Combination of Bag of Words and Point Feature Based Relative Pose Estimation. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2011

    work page 2011

  55. [55]

    Tombari, S

    F. Tombari, S. Salti, and L.D. Stefano. Unique Shape Context for 3d Data Description. In Proceedings of the ACM Workshop on 3D Object Retrieval, 2010

    work page 2010

  56. [56]

    S. Umeyama. Least-squares Estimation of Transformation Parameters Between Two Point Patterns. IEEE Trans. on Pattern Analysis and Machine Intelligence (TPAMI) , 13(4):376–380, 1991

    work page 1991

  57. [57]

    Uy and G

    A. Uy and G. Lee. PointNetVLAD: Deep Point Cloud Based Retrieval for Large-scale Place Recognition. In Proc. of the IEEE Conf. on Computer Vision and Pattern Recognition (CVPR) , 2018

    work page 2018

  58. [58]

    Vizzo, T

    I. Vizzo, T. Guadagnino, B. Mersch, L. Wiesmann, J. Behley, and C. Stachniss. KISS-ICP: In Defense of Point-to-Point ICP – Simple, Accurate, and Robust Registration If Done the Right Way. IEEE Robotics and Automation Letters (RA-L) , 8(2):1029–1036, 2023

    work page 2023

  59. [59]

    Vysotska and C

    O. Vysotska and C. Stachniss. Lazy Data Association For Image Sequences Matching Under Substantial Appearance Changes. IEEE Robotics and Automation Letters (RA-L) , 1(1):213–220, 2016

    work page 2016

  60. [60]

    Vysotska and C

    O. Vysotska and C. Stachniss. Effective Visual Place Recognition Using Multi-Sequence Maps. IEEE Robotics and Automation Letters (RA-L) , 4(2):1730–1736, 2019

    work page 2019

  61. [61]

    Y . Wang, Z. Sun, C. Xu, S. Sarma, J. Yang, and H. Kong. LiDAR Iris for Loop-Closure Detection. In Proc. of the IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS) , 2020

    work page 2020

  62. [62]

    J. Xu, R. Zhang, J. Dou, Y . Zhu, J. Sun, and S. Pu. RPVNet: A Deep and Efficient Range-Point-V oxel Fusion Network for LiDAR Point Cloud Segmentation. In Proc. of the IEEE/CVF Intl. Conf. on Computer Vision (ICCV), 2021

    work page 2021

  63. [63]

    X. Xu, S. Lu, J. Wu, H. Lu, Q. Zhu, Y . Liao, R. Xiong, and Y . Wang. RING++: Roto-Translation Invariant Gram for Global Localization on a Sparse Scan Map. IEEE Trans. on Robotics (TRO) , 39(6):4616–4635, 2023

    work page 2023

  64. [64]

    X. Xu, H. Yin, Z. Chen, Y . Li, Y . Wang, and R. Xiong. DiSCO: Differentiable Scan Context With Orientation. IEEE Robotics and Automation Letters (RA-L) , 6(2):2791–2798, 2021

    work page 2021

  65. [65]

    J. Yang, Q. Zhang, Y . Xiao, and Z. Cao. TOLDI: An Effective and Robust Approach for 3D Local Shape Description. Pattern Recognition, 65:175–187, 2017

    work page 2017

  66. [66]

    H. Yin, X. Xu, S. Lu, X. Chen, R. Xiong, S. Shen, C. Stachniss, and Y . Wang. A Survey on Global LiDAR Localization: Challenges, Advances and Open Problems. Intl. Journal of Computer Vision (IJCV) , 132:1–33, 03 2024

    work page 2024

  67. [67]

    C. Yuan, J. Lin, Z. Liu, H. Wei, X. Hong, and F. Zhang. BTC: A Binary and Triangle Combined Descriptor for 3-D Place Recognition. IEEE Trans. on Robotics (TRO) , 40:1580–1599, 2024

    work page 2024

  68. [68]

    C. Yuan, J. Lin, Z. Zou, X. Hong, and F. Zhang. STD: Stable Triangle Descriptor for 3D place recognition. In Proc. of the IEEE Intl. Conf. on Robotics & Automation (ICRA) , 2023

    work page 2023

  69. [69]

    Zhang, P

    Y . Zhang, P. Shi, and J. Li. LiDAR-Based Place Recognition For Autonomous Driving: A Survey. arXiv preprint , arXiv:2306.10561, 2023

    work page arXiv 2023

  70. [70]

    B. Zhou, Y . He, K. Qian, X. Ma, and X. Li. S4-slam: A real-time 3d lidar slam system for ground/watersurface multi-scene outdoor applications. Autonomous Robots , 45(1):77–98, 2021

    work page 2021

  71. [71]

    Q. Zhou, J. Park, and V . Koltun. Open3D: A modern library for 3D data processing. arXiv preprint, arXiv:1801.09847, 2018

    work page internal anchor Pith review Pith/arXiv arXiv 2018