CaDeX: Learning Canonical Deformation Coordinate Space for Dynamic Surface Representation via Neural Homeomorphism
This addresses the need for efficient and template-independent representations of dynamic surfaces in computer graphics and vision, offering a novel approach with broad applicability.
The paper tackles the problem of representing deformable 3D surfaces by introducing CaDeX, a method that learns a canonical coordinate space via neural homeomorphisms, achieving state-of-the-art performance in modeling human bodies, animal bodies, and articulated objects.
While neural representations for static 3D shapes are widely studied, representations for deformable surfaces are limited to be template-dependent or lack efficiency. We introduce Canonical Deformation Coordinate Space (CaDeX), a unified representation of both shape and nonrigid motion. Our key insight is the factorization of the deformation between frames by continuous bijective canonical maps (homeomorphisms) and their inverses that go through a learned canonical shape. Our novel deformation representation and its implementation are simple, efficient, and guarantee cycle consistency, topology preservation, and, if needed, volume conservation. Our modelling of the learned canonical shapes provides a flexible and stable space for shape prior learning. We demonstrate state-of-the-art performance in modelling a wide range of deformable geometries: human bodies, animal bodies, and articulated objects.