Diffeomorphic Mesh Deformation via Efficient Optimal Transport for Cortical Surface Reconstruction

arXiv:2305.17555v3 Announce Type: replace
Abstract: Mesh deformation plays a pivotal role in many 3D vision tasks including dynamic simulations, rendering, and reconstruction. However, defining an efficient discrepancy between predicted and target meshes remains an open problem. A prevalent approach in current deep learning is the set-based approach which measures the discrepancy between two surfaces by comparing two randomly sampled point-clouds from the two meshes with Chamfer pseudo-distance. Nevertheless, the set-based approach still has limitations such as lacking a theoretical guarantee …

abstract arxiv cs.cv current deep learning dynamic however mesh meshes pivotal rendering role set simulations surface tasks transport type via vision