Laplacian ICP for Progressive Registration of 3D Human Head Meshes
This work addresses efficient 3D registration for human head scans, which is incremental as it builds on classical non-rigid ICP with domain-specific optimizations.
The paper tackles the problem of 3D non-rigid registration for human head meshes by proposing Laplacian ICP, a progressive framework that uses Laplace-Beltrami operator regularization and coarse-to-fine deformation. The result is a method that is robust, requires only a small fraction of the computation time compared to popular classical approaches, and achieves comparable registration performance.
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.