DeePSD: Automatic Deep Skinning And Pose Space Deformation For 3D Garment Animation
This addresses the garment animation problem for 3D graphics and animation industries by offering a simpler, more compatible solution, though it is incremental by focusing solely on animation rather than broader garment editing tasks.
The paper tackles the problem of animating 3D garments with arbitrary topology and complexity by proposing a deep learning model that automatically generates blend weights and blend shapes for compatibility with graphics engines, achieving scalability and compatibility without complex engineering.
We present a novel solution to the garment animation problem through deep learning. Our contribution allows animating any template outfit with arbitrary topology and geometric complexity. Recent works develop models for garment edition, resizing and animation at the same time by leveraging the support body model (encoding garments as body homotopies). This leads to complex engineering solutions that suffer from scalability, applicability and compatibility. By limiting our scope to garment animation only, we are able to propose a simple model that can animate any outfit, independently of its topology, vertex order or connectivity. Our proposed architecture maps outfits to animated 3D models into the standard format for 3D animation (blend weights and blend shapes matrices), automatically providing of compatibility with any graphics engine. We also propose a methodology to complement supervised learning with an unsupervised physically based learning that implicitly solves collisions and enhances cloth quality.