CoBigICP: Robust and Precise Point Set Registration using Correntropy Metrics and Bidirectional Correspondence

TL;DR

CoBigICP introduces a probabilistic ICP variant that combines local geometric bidirectional correspondence with a global correntropy noise model, enhancing robustness and accuracy in point cloud registration.

Contribution

The paper presents CoBigICP, a novel method integrating bidirectional correspondence and correntropy-based noise modeling for improved point cloud registration.

Findings

Outperforms state-of-the-art registration methods

Resists outliers effectively due to correntropy metric

Provides robust and precise registration results

Abstract

In this paper, we propose a novel probabilistic variant of iterative closest point (ICP) dubbed as CoBigICP. The method leverages both local geometrical information and global noise characteristics. Locally, the 3D structure of both target and source clouds are incorporated into the objective function through bidirectional correspondence. Globally, error metric of correntropy is introduced as noise model to resist outliers. Importantly, the close resemblance between normal-distributions transform (NDT) and correntropy is revealed. To ease the minimization step, an on-manifold parameterization of the special Euclidean group is proposed. Extensive experiments validate that CoBigICP outperforms several well-known and state-of-the-art methods.