Laplacian ICP for Progressive Registration of 3D Human Head Meshes

TL;DR

This paper introduces a fast, progressive 3D registration method for human head meshes using Laplace-Beltrami regularization, with a new benchmark for evaluation and promising results.

Contribution

It proposes Laplacian ICP, a highly-efficient non-rigid registration framework with a coarse-to-fine approach and domain-specific correspondence matching.

Findings

Robust registration with significantly reduced computation time

Comparable performance to classical methods

New benchmark and metrics for 3D head registration

Abstract

We present a progressive 3D registration framework that is a highly-efficient variant of classical non-rigid Iterative Closest Points (N-ICP). Since it uses the Laplace-Beltrami operator for deformation regularisation, we view the overall process as Laplacian ICP (L-ICP). This exploits a `small deformation per iteration' assumption and is progressively coarse-to-fine, employing an increasingly flexible deformation model, an increasing number of correspondence sets, and increasingly sophisticated correspondence estimation. Correspondence matching is only permitted within predefined vertex subsets derived from domain-specific feature extractors. Additionally, we present a new benchmark and a pair of evaluation metrics for 3D non-rigid registration, based on annotation transfer. We use this to evaluate our framework on a publicly-available dataset of 3D human head scans (Headspace). The method is robust and only requires a small fraction of the computation time compared to the most popular classical approach, yet has comparable registration performance.