Learning‐Based Animation of Clothing for Virtual Try‐On

TLDR

The study proposes a learning‑based clothing animation method that separates global garment fit from local wrinkles, using a recurrent neural network to regress wrinkles and achieve plausible nonlinear effects for efficient virtual try‑on. The approach builds a database of physically‑based simulations for various body shapes and animations, trains a learning‑based model of cloth drape and wrinkles conditioned on body shape and dynamics, and evaluates the results qualitatively and quantitatively. The recurrent neural network yields highly plausible nonlinear wrinkle effects without blending artifacts, enabling dynamic virtual try‑on animations in a few milliseconds even for garments with thousands of triangles.

Abstract

Abstract This paper presents a learning‐based clothing animation method for highly efficient virtual try‐on simulation. Given a garment, we preprocess a rich database of physically‐based dressed character simulations, for multiple body shapes and animations. Then, using this database, we train a learning‐based model of cloth drape and wrinkles, as a function of body shape and dynamics. We propose a model that separates global garment fit, due to body shape, from local garment wrinkles, due to both pose dynamics and body shape. We use a recurrent neural network to regress garment wrinkles, and we achieve highly plausible nonlinear effects, in contrast to the blending artifacts suffered by previous methods. At runtime, dynamic virtual try‐on animations are produced in just a few milliseconds for garments with thousands of triangles. We show qualitative and quantitative analysis of results.