A curated list of papers about point cloud registration inspired by awesome point cloud analysis

You will be very welcome to make PR and contribute!! 😄

lf.: local features for registration   |   corr.: register with putative correspondences   |  

est.: direct estimation   |   dat.: datasets   |  

opt.: optimization   |   oth.: other

• [CGF] SUPER 4PCS: Fast Global Pointcloud Registration via Smart Indexing. [code] [est. oth.] 🔥 ⭐

• [CGF] Sparse Iterative Closest Point. [code] [est. opt.] 🔥 ⭐

• [IJRR] Challenging data sets for point cloud registration algorithms. [code] [dat.] ⭐

• [CVPR] 3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions. [code] [lf. dat.] 🔥 ⭐

• [CVPR] 3D Point Cloud Registration for Localization using a Deep Neural Network Auto-Encoder. [code] [est.]

• [ICCV] Learning Compact Geometric Features. [code] [est. dat.]

• [ECCV] 3DFeat-Net: Weakly Supervised Local 3D Features for Point Cloud Registration [code] [lf.]

• [ECCV] PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors [lf.]

• [CVPR] PPFNet: Global Context Aware Local Features for Robust 3D Point Matching [lf.]

• [CVPR] End-to-end learning of keypoint detector and descriptor for pose invariant 3D matching [lf.]

• [CVPR] Inverse Composition Discriminative Optimization for Point Cloud Registration [opt.]

• [CVPR] PointNetLK: Robust & Efficient Point Cloud Registration using PointNet. [code] [est.] 🔥

• [CVPR] DeepMapping: Unsupervised Map Estimation From Multiple Point Clouds. [code] [est. opt.] 🔥

• [CVPR] The Perfect Match: 3D Point Cloud Matching with Smoothed Densities [code] [lf.]

• [CVPR] L3 -Net: Towards Learning based LiDAR Localization for Autonomous Driving [lf. est.]

• [CVPR] SDRSAC: Semidefinite-Based Randomized Approach for Robust Point Cloud Registration without Correspondences [code] [ est. opt.]

• [CVPR] 3D Local Features for Direct Pairwise Registration [lf. est.]

• [CVPR] 3D Point Capsule Networks [code] [lf.] 🔥

• [CVPR] GFrames: Gradient-Based Local Reference Frame for 3D Shape Matching [oth.]

• [CVPR] FilterReg: Robust and Efficient Probabilistic Point-Set Registration using Gaussian Filter and Twist Parameterization [code] [opt. oth.]

• [ICCV] DeepVCP: An End-to-End Deep Neural Network for Point Cloud Registration [est. lf.]

• [ICCV] Deep Closest Point: Learning Representations for Point Cloud Registration [code] [est.]

• [ICCV] USIP: Unsupervised Stable Interest Point Detection from 3D Point Clouds. [code] [lf.]

• [ICCV] Learning an Effective Equivariant 3D Descriptor Without Supervision. [lf.]

• [ICCV] Fully Convolutional Geometric Features. [code][lf.]

• [ICRA] Robust low-overlap 3-D point cloud registration for outlier rejection [est. opt. corr. oth.]

• [ICRA] CELLO-3D: Estimating the Covariance of ICP in the Real World [est. opt. oth.]

• [NeurIPS] PRNet: Self-Supervised Learning for Partial-to-Partial Registration [est.]

• [TOG] A Symmetric Objective Function for ICP [est. opt.]

• [ARXIV] 3DRegNet: A Deep Neural Network for 3D Point Registration [est.]

• [ARXIV] Iterative Matching Point [est.]

• [ARXIV] PCRNet: Point Cloud Registration Network using PointNet Encoding [code] [est.]

• [CVPR] End-to-End Learning Local Multi-view Descriptors for 3D Point Clouds [code] [lf.]
• [CVPR] D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features [code] [lf.]
• [CVPR] 3DRegNet: A Deep Neural Network for 3D Point Registration [code] [corr.]
• [CVPR] High-Dimensional Convolutional Networks for Geometric Pattern Recognition [code] [corr.]
• [CVPR] Deep Global Registration [code] [lf. corr.]
• [CVPR] Learning multiview 3D point cloud registration [code] [corr. oth.]
• [CVPR] RPM-Net: Robust Point Matching using Learned Features [code] [est.]
• [CVPR] Feature-metric Registration: A Fast Semi-supervised Approach for Robust Point Cloud Registration without Correspondences [code] [opt. est. oth.]
• [CVPR] PointGMM: a Neural GMM Network for Point Clouds [est. oth.]
• [ECCV] DeepGMR: Learning Latent Gaussian Mixture Models for Registration [code] [est. corr.]
• [ARXIV] TEASER: Fast and Certifiable Point Cloud Registration. [code] [corr. opt.]

• [ICRA] A Single Correspondence Is Enough: Robust Global Registration to Avoid Degeneracy in Urban Environments [code] [corr. opt.] 🔥

• [Measurement,Elsever] Benchmark of multi-view Terrestrial Laser Scanning Point Cloud data registration algorithms [code] [est. opt. oth.]
• [ICRA] Segregator: Global Point Cloud Registration with Semantic and Geometric Cues [code] [corr. opt.] 🔥
• [TASE] G3Reg: Pyramid Graph-based Global Registration using Gaussian Ellipsoid Model [code] [corr. opt.] 🔥