awesome-point-cloud-place-recognition

For anyone who wants to do research about 3D point cloud based place recognition / loop closure detection. Thanks.

- Recent papers (from 2016)

Keywords

out.: outdoor | ind.: indoor
pc.: point cloud | img.: image | rad.: radar
pos.: pose

Statistics: 🔥 code is available & stars >= 100 | ⭐ citation >= 50

2016

[IROS] M2DP: A novel 3D point cloud descriptor and its application in loop closure detection. [matlab] [python] [out. pc.]

2017

2018

[IROS] Scan Context: Egocentric Spatial Descriptor for Place Recognition Within 3D Point Cloud Map. [code] [out. pc. pos.] 🔥 ⭐
[IV] LocNet: Global Localization in 3D Point Clouds for Mobile Vehicles. [code] [out. pc.]
[RSS] SegMap: 3D Segment Mapping using Data-Driven Descriptors. [code] [out. pc.] 🔥 ⭐
[CVPR] PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition. [code] [out. pc.] 🔥 ⭐

2019

[RA-L] 1-Day Learning, 1-Year Localization: Long-Term LiDAR Localization Using Scan Context Image. [code] [out. pc. pos.] 🔥
[RA-L] Local Descriptor for Robust Place Recognition Using LiDAR Intensity. [out. pc.]
[RA-L] Learning to See the Wood for the Trees: Deep Laser Localization in Urban and Natural Environments on a CPU. [out. pc.]
[IROS] OREOS: Oriented Recognition of 3D Point Clouds in Outdoor Scenarios. [out. pc. pos.]
[IROS] Semantically Assisted Loop Closure in SLAM Using NDT Histograms. [out. pc.]
[IROS] SeqLPD: Sequence Matching Enhanced Loop-Closure Detection Based on Large-Scale Point Cloud Description for Self-Driving Vehicles. [out. pc.]
[ICCV] LPD-Net: 3D Point Cloud Learning for Large-Scale Place Recognition and Environment Analysis. [code] [out. pc.]
[RSS] OverlapNet - Loop Closing for 3D LiDAR-based SLAM. [code] [out. pc. pos.] 🔥
[IJRR] SegMap: Segment-based mapping and localization using data-driven descriptors. [code] [out. pc.] 🔥
[Sensors] Have I Seen This Place Before? A Fast and Robust Loop Detection and Correction Method for 3D Lidar SLAM. [out. pc.]
[TIE] Season-Invariant and Viewpoint-Tolerant LiDAR Place Recognition in GPS-Denied Environments. [out. pc.]
[arXiv] A fast, complete, point cloud based loop closure for LiDAR odometry and mapping. [code] [out. pc.] 🔥

2020

[IROS] Intensity Scan Context: Coding Intensity and Geometry Relations for Loop Closure Detection. [code] [out. pc. pos.] 🔥
[IROS] Semantic Graph Based Place Recognition for 3D Point Clouds. [code] [out. pc.]
[IROS] LiDAR Iris for Loop-Closure Detection. [code] [out. pc. pos.]
[IROS] Seed: A Segmentation-Based Egocentric 3D Point Cloud Descriptor for Loop Closure Detection. [out. pc.]
[IROS] GOSMatch: Graph-of-Semantics Matching for Detecting Loop Closures in 3D LiDAR data. [code] [out. pc.]
[IROS] Gaussian Process Gradient Maps for Loop-Closure Detection in Unstructured Planetary Environments. [out. pc.]
[IROS] SpoxelNet: Spherical Voxel-based Deep Place Recognition for 3D Point Clouds of Crowded Indoor Spaces. [ind. pc.]
[IROS] Voxel-Based Representation Learning for Place Recognition Based on 3D Point Clouds. [out. pc.]
[IROS] SeqSphereVLAD: Sequence Matching Enhanced Orientation-invariant Place Recognition. [out. pc.]
[AR] Loop detection for 3D LiDAR SLAM using segment-group matching. [out. pc.]
[TIE] Fast Sequence-matching Enhanced Viewpoint-invariant 3D Place Recognition. [out. pc.]
[Sensors] Large-Scale Place Recognition Based on Camera-LiDAR Fused Descriptor. [out. pc. img.]
[ICMR] DAGC: Employing Dual Attention and Graph Convolution for Point Cloud based Place Recognition. [out. pc.]
[ICARCV] Comparison of camera-based and 3D LiDAR-based place recognition across weather conditions. [out. pc. img.]
[arXiv] vLPD-Net: A Registration-aided Domain Adaptation Network for 3D Point Cloud Based Place Recognition. [out. pc. img.]
[arXiv] PIC-Net: Point Cloud and Image Collaboration Network for Large-Scale Place Recognition. [out. pc. img.]

2021

[RA-L] DiSCO: Differentiable Scan Context With Orientation. [code] [out. pc. pos.]
[ICRA] Locus: LiDAR-based Place Recognition using Spatiotemporal Higher-Order Pooling. [code] [out. pc.]
[ICRA] NDT-Transformer: Large-Scale 3D Point Cloud Localisation using the Normal Distribution Transform Representation. [code] [out. pc.]
[ICRA] Robust Place Recognition using an Imaging Lidar. [code] [out. pc.] 🔥
[RA-L] LiPMatch: LiDAR Point Cloud Plane based Loop-Closure. [code] [out. pc.]
[RA-L] FusionVLAD: A Multi-View Deep Fusion Networks for Viewpoint-Free 3D Place Recognition. [out. pc.]
[WACV] MinkLoc3D: Point Cloud Based Large-Scale Place Recognition. [code] [out. pc.]
[IJCNN] MinkLoc++: Lidar and Monocular Image Fusion for Place Recognition. [code] [out. pc. img.]
[ICCCR] Weighted Scan Context: Global Descriptor with Sparse Height Feature for Loop Closure Detection. [out. pc. pos.]
[Frontiers] Radar-to-Lidar: Heterogeneous Place Recognition via Joint Learning. [code] [out. pc. rad.]
[arXiv] Efficient 3D Point Cloud Feature Learning for Large-Scale Place Recognition. [code] [out. pc.]
[arXiv] Have I been here before? Learning to Close the Loop with LiDAR Data in Graph-Based SLAM. [out. pc.]
[arXiv] LCDNet: Deep Loop Closure Detection for LiDAR SLAM based on Unbalanced Optimal Transport. [code] [out. pc. pos.]
[IROS] SSC: Semantic Scan Context for Large-Scale Place Recognition. [code] [out. pc. pos.]