AI RESEARCH

DeepShapeMatchingKit: Accelerated Functional Map Solver and Shape Matching Pipelines Revisited

arXiv CS.CV

ArXi:2604.10377v1 Announce Type: new Deep functional maps, leveraging learned feature extractors and spectral correspondence solvers, are fundamental to non-rigid 3D shape matching. Based on an analysis of open-source implementations, we find that standard functional map implementations solve k independent linear systems serially, which is a computational bottleneck at higher spectral resolution. We thus propose a vectorized reformulation that solves all systems in a single kernel call, achieving up to a 33x speedup while preserving the exact solution.