An approach to robust ICP initialization

Alexander Kolpakov, Michael Werman

Research output: Contribution to journalArticlepeer-review


In this note, we propose an approach to initialize the Iterative Closest Point (ICP) algorithm to match unlabelled point clouds related by rigid transformations. The method is based on matching the ellipsoids defined by the points' covariance matrices and then testing the various principal half–axes matchings that differ by elements of a finite reflection group. We derive bounds on the robustness of our approach to noise and numerical experiments confirm our theoretical findings.

Original languageAmerican English
Pages (from-to)12685-12691
Number of pages7
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Issue number10
StatePublished - 2023

Bibliographical note

Publisher Copyright:


  • Ellipsoids
  • Iterative closest point algorithm
  • Noise measurement
  • Point cloud compression
  • Random variables
  • Symmetric matrices
  • Three-dimensional displays


Dive into the research topics of 'An approach to robust ICP initialization'. Together they form a unique fingerprint.

Cite this