Learning Canonical Embedding for Non-rigid Shape Matching

Abhishek Sharma, Maks Ovsjanikov

Keywords: Deep Functional Maps, Symmetry group, Point Clouds, Linear Transformation, Canonical 3D shape Embedding

Abstract: This paper provides a novel framework that learns canonical embeddings for non-rigid shape matching. In contrast to prior work in this direction, our framework is trained end-to-end and thus avoids instabilities and constraints associated with the commonly-used Laplace-Beltrami basis or sequential optimization schemes. On multiple datasets, we demonstrate that learning self symmetry maps with a deep functional map projects 3D shapes into a low dimensional canonical embedding that facilitates non-rigid shape correspondence via a simple nearest neighbor search. Our framework outperforms multiple recent learning based methods on FAUST and SHREC benchmarks while being computationally cheaper, data-efficient, and robust.

One-sentence Summary: learn an embedding of each shape that would make the given self-symmetry map linear in some higher-dimensional space.