Linear Least-Squares Optimization for Point-to-Plane ICP Surface Registration

By Kok-Lim Low. 2004

568 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/725c7ce8224b56caffeb2c5cd41a04582c5c3563

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Modeling Kinect Sensor Noise for Improved 3D Reconstruction and Tracking (2012, background): “The point-plane energy function is linearized by assuming small angle changes from one pose to the next [6].”
Geometric distortion metrics for point cloud compression (2017, background): “, Iterative Closest Point (ICP) algorithms [5, 6] for faster convergence.”
A comprehensive survey on point cloud registration (2021, methodology): “Low et al.[63] propose a linear approximation of the rotation matrix and estimate the transformation using SVD.”
KinectFusion: real-time 3D reconstruction and interaction using a moving depth camera (2011, methodology): “We make a linear approximation to solve this system, by assuming only an incremental transformation occurs between frames [3, 17].”
State of the Art on 3D Reconstruction with RGB‐D Cameras (2018, methodology): “Tracking is based on the point-to-plane ICP described by [Low04a].”
Real-time 3D reconstruction at scale using voxel hashing (2013, methodology): “The point-plane energy function is linearized [Low 2004] on the GPU to a 6 × 6 matrix using a parallel reduction and solved via Singular Value Decomposition on the CPU.”
Privacy Preserving Image Queries for Camera Localization (2019, methodology): “Our privacy preserving localization approach relies on a client-server architecture, where the client ﬁrst extracts lo-cal 2D features from the query image ( e.g ., SIFT [39]) and then sends them to the server for computing the camera pose w.r.t. a pre-computed 3D point cloud.”
IMLS-SLAM: Scan-to-Model Matching Based on 3D Data (2018, background): “Like in ICP pointto-plane, we can now make the small angle assumption on R to get a linear least-square optimization problem that can be solved efficiently (more details in the technical report [16]).”