A Repulsive Force Unit for Garment Collision Handling in Neural Networks

TL;DR

This paper introduces ReFU, a neural network layer that effectively handles garment-body collisions in 3D deformation prediction, improving collision reduction and detail preservation.

Contribution

ReFU is a novel, differentiable collision handling layer based on SDF that can be integrated into existing networks for better garment deformation predictions.

Findings

ReFU significantly reduces garment-body collisions.

ReFU preserves geometric details better than prior methods.

ReFU is adaptable to various network architectures.

Abstract

Despite recent success, deep learning-based methods for predicting 3D garment deformation under body motion suffer from interpenetration problems between the garment and the body. To address this problem, we propose a novel collision handling neural network layer called Repulsive Force Unit (ReFU). Based on the signed distance function (SDF) of the underlying body and the current garment vertex positions, ReFU predicts the per-vertex offsets that push any interpenetrating vertex to a collision-free configuration while preserving the fine geometric details. We show that ReFU is differentiable with trainable parameters and can be integrated into different network backbones that predict 3D garment deformations. Our experiments show that ReFU significantly reduces the number of collisions between the body and the garment and better preserves geometric details compared to prior methods based on collision loss or post-processing optimization.