Deep Global Registration
Christopher Choy, Wei Dong, Vladlen Koltun · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · 2020
We present Deep Global Registration, a differentiable framework for pairwise registration of real-world 3D scans. Deep global registration is based on three modules: a 6-dimensional convolutional network for correspondence confidence prediction, a differentiable Weighted Procrustes algorithm for closed-form pose estimation, and a robust gradient-based SE(3) optimizer for pose refinement. Experiments demonstrate that our approach outperforms state-of-the-art methods, both learning-based and classical, on real-world data.
Cited by 548 works. doi:10.1109/cvpr42600.2020.00259 Open-access PDF
Authors
References
- A method for registration of 3-D shapes
- An Iterative Image Registration Technique with an Application to Stereo Vision
- Are we ready for autonomous driving? The KITTI vision benchmark suite
- Using spin images for efficient object recognition in cluttered 3D scenes
- Iterative point matching for registration of free-form curves and surfaces
- ORB-SLAM: A Versatile and Accurate Monocular SLAM System
- Generalized Procrustes Analysis
- Fast Point Feature Histograms (FPFH) for 3D registration
- ORB-SLAM: A Versatile and Accurate Monocular SLAM System
- G2o: A general framework for graph optimization