mlsanityData to 14 Oct 2026

CorrI2P: Deep Image-to-Point Cloud Registration via Dense Correspondence

arXiv 2207.05483 ↗First posted 12 Jul 2022cs.CV cs.MMcited by 44

Not on the map. Why?

Abstract (arXiv)

Motivated by the intuition that the critical step of localizing a 2D image in the corresponding 3D point cloud is establishing 2D-3D correspondence between them, we propose the first feature-based dense correspondence framework for addressing the image-to-point cloud registration problem, dubbed CorrI2P, which consists of three modules, i.e., feature embedding, symmetric overlapping region detection, and pose estimation through the established correspondence. Specifically, given a pair of a 2D image and a 3D point cloud, we first transform them into high-dimensional feature space and feed the resulting features into a symmetric overlapping region detector to determine the region where the image and point cloud overlap each other. Then we use the features of the overlapping regions to establish the 2D-3D correspondence before running EPnP within RANSAC to estimate the camera's pose. Experimental results on KITTI and NuScenes datasets show that our CorrI2P outperforms state-of-the-art image-to-point cloud registration methods significantly. We will make the code publicly available.

Key facts. Cited by 44 papers in the map (data through 14 Oct 2026). Builds on 28 of its 64 references that are in the map. First posted on arXiv 12 Jul 2022.

Builds on

28 of 64 references.

References · 28

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  4. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  5. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  6. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  7. 20163DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructionscited by 336
  8. 2019Deep Closest Point: Learning Representations for Point Cloud Registrationcited by 321
  9. 2020PREDATOR: Registration of 3D Point Clouds with Low Overlapcited by 290
  10. 2019PointNetLK: Robust & Efficient Point Cloud Registration using PointNetcited by 233
  11. 2018SO-Net: Self-Organizing Network for Point Cloud Analysiscited by 210
  12. 2020RPM-Net: Robust Point Matching using Learned Featurescited by 182
16 more
  1. 2018PPFNet: Global Context Aware Local Features for Robust 3D Point Matchingcited by 181
  2. 2020D3Feat: Joint Learning of Dense Detection and Description of 3D Local Featurescited by 177
  3. 2018The Perfect Match: 3D Point Cloud Matching with Smoothed Densitiescited by 169
  4. 2021PointDSC: Robust Point Cloud Registration using Deep Spatial Consistencycited by 142
  5. 2016Image-based localization using LSTMs for structured feature correlationcited by 136
  6. 2017Deep Learning Features at Scale for Visual Place Recognitioncited by 96
  7. 2019PCRNet: Point Cloud Registration Network using PointNet Encodingcited by 63
  8. 20192D3D-MatchNet: Learning to Match Keypoints Across 2D Image and 3D Point Cloudcited by 62
  9. 2021DeepI2P: Image-to-Point Cloud Registration via Deep Classificationcited by 59
  10. 2017VidLoc: A Deep Spatio-Temporal Model for 6-DoF Video-Clip Relocalizationcited by 58
  11. 2021HRegNet: A Hierarchical Network for Large-scale Outdoor LiDAR Point Cloud Registrationcited by 48
  12. 2020CorrNet3D: Unsupervised End-to-end Learning of Dense Correspondence for 3D Point Cloudscited by 47
  13. 2017Learning Compact Geometric Featurescited by 45
  14. 2020Monocular Camera Localization in Prior LiDAR Maps with 2D-3D Line Correspondencescited by 23
  15. 2020Recurrent Multi-view Alignment Network for Unsupervised Surface Registrationcited by 19
  16. 2020Learning 2D-3D Correspondences To Solve The Blind Perspective-n-Point Problemcited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 7; 2024: 12; 2025: 13; 2026: 12.

44 papers.

Sort:

By area:

  1. 2023Differentiable Registration of Images and LiDAR Point Clouds with VoxelPoint-to-Pixel Matchingcited by 37
  2. 2023FreeReg: Image-to-Point Cloud Registration Leveraging Pretrained Diffusion Models and Monocular Depth Estimatorscited by 31
  3. 2025Bridge 2D-3D: Uncertainty-aware Hierarchical Registration Network with Domain Alignmentcited by 14
  4. 2023CoFiI2P: Coarse-to-Fine Correspondences for Image-to-Point Cloud Registrationcited by 12
  5. 2023End-to-end 2D-3D Registration between Image and LiDAR Point Cloud for Vehicle Localizationcited by 9
  6. 2025CA-I2P: Channel-Adaptive Registration Network with Global Optimal Selectioncited by 8
  7. 2025Diff2I2P: Differentiable Image-to-Point Cloud Registration with Diffusion Priorcited by 8
  8. 2024Self-supervised Learning of LiDAR 3D Point Clouds via 2D-3D Neural Calibrationcited by 7
  9. 2025Adaptive Agent Selection and Interaction Network for Image-to-point cloud Registrationcited by 6
  10. 2024VXP: Voxel-Cross-Pixel Large-scale Image-LiDAR Place Recognitioncited by 5
  11. 2024Point Tree Transformer for Point Cloud Registrationcited by 5
  12. 2024EM-MIAs: Enhancing Membership Inference Attacks in Large Language Models through Ensemble Modelingcited by 5
32 more
  1. 2026IGASA: Integrated Geometry-Aware and Skip-Attention Modules for Enhanced Point Cloud Registrationcited by 5
  2. 2023Quantity-Aware Coarse-to-Fine Correspondence for Image-to-Point Cloud Registrationcited by 4
  3. 2024CMR-Agent: Learning a Cross-Modal Agent for Iterative Image-to-Point Cloud Registrationcited by 3
  4. 2025EdgeRegNet: Edge Feature-based Multimodal Registration Network between Images and LiDAR Point Cloudscited by 3
  5. 2024SOLVR: Submap Oriented LiDAR-Visual Re-Localisationcited by 2
  6. 2024TrafficLoc: Localizing Traffic Surveillance Cameras in 3D Scenescited by 2
  7. 2023SDFReg: Learning Signed Distance Functions for Point Cloud Registrationcited by 1
  8. 2024EEPNet: Efficient Edge Pixel-based Matching Network for Cross-Modal Dynamic Registration between LiDAR and Cameracited by 1
  9. 2024Iterative Camera-LiDAR Extrinsic Optimization via Surrogate Diffusioncited by 1
  10. 2025Object Affordance Recognition and Grounding via Multi-scale Cross-modal Representation Learningcited by 1
  11. 2025Self-Supervised Cross-Modal Learning for Image-to-Point Cloud Registrationcited by 1
  12. 2026GLASS: Geometry-aware Local Alignment and Structure Synchronization Network for 2D-3D Registrationcited by 1
  13. 2026Hg-I2P: Bridging Modalities for Generalizable Image-to-Point-Cloud Registration via Heterogeneous Graphscited by 1
  14. 2023VioLA: Aligning Videos to 2D LiDAR Scanscited by 0
  15. 2024Towards Zero-Shot Interpretable Human Recognition: A 2D-3D Registration Frameworkcited by 0
  16. 2024MaFreeI2P: A Matching-Free Image-to-Point Cloud Registration Paradigm with Active Camera Pose Retrievalcited by 0
  17. 2024TFCT-I2P: Three stream fusion network with color aware transformer for image-to-point cloud registrationcited by 0
  18. 2025EEPNet-V2: Patch-to-Pixel Solution for Efficient Cross-Modal Registration between LiDAR Point Cloud and Camera Imagecited by 0
  19. 2025TIGeR: Text-Instructed Generation and Refinement for Template-Free Hand-Object Interactioncited by 0
  20. 2025Iterative Camera-LiDAR Extrinsic Optimization via Surrogate Diffusioncited by 0
  21. 2025RAG-6DPose: Retrieval-Augmented 6D Pose Estimation via Leveraging CAD as Knowledge Basecited by 0
  22. 2025Single-Frame Point-Pixel Registration via Supervised Cross-Modal Feature Matchingcited by 0
  23. 2025CRISTAL: Real-time Camera Registration in Static LiDAR Scans using Neural Renderingcited by 0
  24. 2026Native-Domain Cross-Attention for Camera-LiDAR Extrinsic Calibration Under Large Initial Perturbationscited by 0
  25. 2026FS-I2P:A Hierarchical Focus-Sweep Registration Network with Dynamically Allocated Depthcited by 0
  26. 2026Image-to-Point Cloud Registration Made Easy with Rectified Flow-based LiDAR Upsamplingcited by 0
  27. 2026RECO: Region-Aware Compensation for Extrinsic Perturbations in Roadside 3D Detectioncited by 0
  28. 2026CalibBEV: LiDAR-Camera Calibration via BEV Alignmentcited by 0
  29. 2026DRS-VPT: Directly Relocalizing in a Scan with Vision Point Transformerscited by 0
  30. 2026Mask 2D-3D: Adaptive Dual-Masked Autoencoder Network for Image-to-Point Cloud Registrationcited by 0
  31. 2026ZIL: Zero-shot Image-to-LiDAR Registrationcited by 0
  32. 2026OCA: ODE-Driven Cross-Attention for Image-to-Point-Cloud Registrationcited by 0