mlsanityData to 7 Oct 2026

MinkLoc3D-SI: 3D LiDAR place recognition with sparse convolutions, spherical coordinates, and intensity

arXiv 2112.06539 ↗First posted 13 Dec 2021cs.RO cs.CVcited by 12

Not on the map. Why?

Abstract (arXiv)

The 3D LiDAR place recognition aims to estimate a coarse localization in a previously seen environment based on a single scan from a rotating 3D LiDAR sensor. The existing solutions to this problem include hand-crafted point cloud descriptors (e.g., ScanContext, M2DP, LiDAR IRIS) and deep learning-based solutions (e.g., PointNetVLAD, PCAN, LPDNet, DAGC, MinkLoc3D), which are often only evaluated on accumulated 2D scans from the Oxford RobotCar dataset. We introduce MinkLoc3D-SI, a sparse convolution-based solution that utilizes spherical coordinates of 3D points and processes the intensity of 3D LiDAR measurements, improving the performance when a single 3D LiDAR scan is used. Our method integrates the improvements typical for hand-crafted descriptors (like ScanContext) with the most efficient 3D sparse convolutions (MinkLoc3D). Our experiments show improved results on single scans from 3D LiDARs (USyd Campus dataset) and great generalization ability (KITTI dataset). Using intensity information on accumulated 2D scans (RobotCar Intensity dataset) improves the performance, even though spherical representation doesn't produce a noticeable improvement. As a result, MinkLoc3D-SI is suited for single scans obtained from a 3D LiDAR, making it applicable in autonomous vehicles.

Key facts. Cited by 12 papers in the map (data through 7 Oct 2026). Builds on 34 of its 43 references that are in the map. First posted on arXiv 13 Dec 2021.

Builds on

34 of 43 references.

References · 34

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  2. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  4. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,707
  5. 2017VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detectioncited by 1,323
  6. 20194D Spatio-Temporal ConvNets: Minkowski Convolutional Neural Networkscited by 1,201
  7. 2015NetVLAD: CNN architecture for weakly supervised place recognitioncited by 935
  8. 2017In Defense of the Triplet Loss for Person Re-Identificationcited by 784
  9. 2019ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networkscited by 572
  10. 2017Fine-tuning CNN Image Retrieval with No Human Annotationcited by 497
  11. 2020Searching Efficient 3D Architectures with Sparse Point-Voxel Convolutioncited by 422
  12. 2018PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognitioncited by 197
22 more
  1. 2020MinkLoc3D: Point Cloud Based Large-Scale Place Recognitioncited by 114
  2. 2019Fast Sparse ConvNetscited by 86
  3. 2021OverlapNet: Loop Closing for LiDAR-based SLAMcited by 72
  4. 2021Scan Context++: Structural Place Recognition Robust to Rotation and Lateral Variations in Urban Environmentscited by 72
  5. 2019PCAN: 3D Attention Map Learning Using Contextual Information for Point Cloud Based Retrievalcited by 68
  6. 2020SOE-Net: A Self-Attention and Orientation Encoding Network for Point Cloud based Place Recognitioncited by 66
  7. 2015Learning Descriptors for Object Recognition and 3D Pose Estimationcited by 60
  8. 2019SegMap: Segment-based mapping and localization using data-driven descriptorscited by 42
  9. 2020DH3D: Deep Hierarchical 3D Descriptors for Robust Large-Scale 6DoF Relocalizationcited by 42
  10. 2021MinkLoc++: Lidar and Monocular Image Fusion for Place Recognitioncited by 41
  11. 2020Locus: LiDAR-based Place Recognition using Spatiotemporal Higher-Order Poolingcited by 38
  12. 2021NDT-Transformer: Large-Scale 3D Point Cloud Localisation using the Normal Distribution Transform Representationcited by 38
  13. 2020Intensity Scan Context: Coding Intensity and Geometry Relations for Loop Closure Detectioncited by 37
  14. 2021TransLoc3D : Point Cloud based Large-scale Place Recognition using Adaptive Receptive Fieldscited by 29
  15. 2020DiSCO: Differentiable Scan Context with Orientationcited by 28
  16. 2020Learning to Localize Using a LiDAR Intensity Mapcited by 28
  17. 2021BVMatch: Lidar-based Place Recognition Using Bird's-eye View Imagescited by 25
  18. 2021SSC: Semantic Scan Context for Large-Scale Place Recognitioncited by 24
  19. 2021Efficient 3D Point Cloud Feature Learning for Large-Scale Place Recognitioncited by 21
  20. 2018Local Descriptor for Robust Place Recognition using LiDAR Intensitycited by 19
  21. 2020CORAL: Colored structural representation for bi-modal place recognitioncited by 18
  22. 2021On the descriptive power of LiDAR intensity images for segment-based loop closing in 3-D SLAMcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 2; 2024: 4; 2025: 2; 2026: 1.

12 papers.

  1. 2022RING++: Roto-translation Invariant Gram for Global Localization on a Sparse Scan Mapcited by 29
  2. 2024General Place Recognition Survey: Towards Real-World Autonomycited by 23
  3. 2022CASSPR: Cross Attention Single Scan Place Recognitioncited by 22
  4. 2022General Place Recognition Survey: Towards the Real-world Autonomy Agecited by 11
  5. 2023CrossLoc3D: Aerial-Ground Cross-Source 3D Place Recognitioncited by 10
  6. 2024MinkUNeXt: Point Cloud-based Large-scale Place Recognition using 3D Sparse Convolutionscited by 6
  7. 2024VXP: Voxel-Cross-Pixel Large-scale Image-LiDAR Place Recognitioncited by 4
  8. 2023Pose-Graph Attentional Graph Neural Network for Lidar Place Recognitioncited by 2
  9. 2024PRFusion: Toward Effective and Robust Multi-Modal Place Recognition with Image and Point Cloud Fusioncited by 2
  10. 2025MinkUNeXt-SI: Improving point cloud-based place recognition including spherical coordinates and LiDAR intensitycited by 1
  11. 2026Advanced techniques and applications of LiDAR Place Recognition in Agricultural Environments: A Comprehensive Surveycited by 1
  12. 2025Towards classification-based representation learning for place recognition on LiDAR scanscited by 0