Geodesic squared exponential kernel for non-rigid shape registration
This work addresses non-rigid shape registration for 3D modeling, particularly for faces, with incremental improvements in kernel design and robustness.
The paper tackles non-rigid registration of 3D scans by proposing a geodesic squared exponential kernel for Gaussian Process Morphable Models and a modified loss function for non-rigid ICP, showing significantly better performance than state-of-the-art kernels on the FaceWarehouse dataset across all 20 expressions.
This work addresses the problem of non-rigid registration of 3D scans, which is at the core of shape modeling techniques. Firstly, we propose a new kernel based on geodesic distances for the Gaussian Process Morphable Models (GPMMs) framework. The use of geodesic distances into the kernel makes it more adapted to the topological and geometric characteristics of the surface and leads to more realistic deformations around holes and curved areas. Since the kernel possesses hyperparameters we have optimized them for the task of face registration on the FaceWarehouse dataset. We show that the Geodesic squared exponential kernel performs significantly better than state of the art kernels for the task of face registration on all the 20 expressions of the FaceWarehouse dataset. Secondly, we propose a modification of the loss function used in the non-rigid ICP registration algorithm, that allows to weight the correspondences according to the confidence given to them. As a use case, we show that we can make the registration more robust to outliers in the 3D scans, such as non-skin parts.