Neural Intrinsic Embedding for Non-rigid Point Cloud Matching
This addresses the challenge of non-rigid point cloud matching for 3D sensing applications, offering a weakly-supervised approach that reduces reliance on extensive supervision and complex geometric inputs.
The paper tackles the problem of establishing correspondences between point clouds from deformable shapes by proposing Neural Intrinsic Embedding (NIE) to embed vertices into a high-dimensional space that respects intrinsic structure, and it shows that the framework performs on par with or better than state-of-the-art baselines while requiring less supervision and structural input.
As a primitive 3D data representation, point clouds are prevailing in 3D sensing, yet short of intrinsic structural information of the underlying objects. Such discrepancy poses great challenges on directly establishing correspondences between point clouds sampled from deformable shapes. In light of this, we propose Neural Intrinsic Embedding (NIE) to embed each vertex into a high-dimensional space in a way that respects the intrinsic structure. Based upon NIE, we further present a weakly-supervised learning framework for non-rigid point cloud registration. Unlike the prior works, we do not require expansive and sensitive off-line basis construction (e.g., eigen-decomposition of Laplacians), nor do we require ground-truth correspondence labels for supervision. We empirically show that our framework performs on par with or even better than the state-of-the-art baselines, which generally require more supervision and/or more structural geometric input.