DSNet: Dynamic Skin Deformation Prediction by Recurrent Neural Network

TL;DR

This paper introduces DSNet, a recurrent neural network that predicts real-time, personalized skin deformation dynamics from mesh data, offering a fast and realistic alternative to traditional physics-based simulations.

Contribution

The paper presents a novel learning-based approach using RNNs for dynamic skin deformation prediction, enabling real-time, personalized, high-quality skin animation.

Findings

Achieves realistic skin dynamics in real-time

Reduces computational time significantly

Maintains high prediction quality compared to state-of-the-art methods

Abstract

Skin dynamics contributes to the enriched realism of human body models in rendered scenes. Traditional methods rely on physics-based simulations to accurately reproduce the dynamic behavior of soft tissues. Due to the model complexity and thus the heavy computation, however, they do not directly offer practical solutions to domains where real-time performance is desirable. The quality shapes obtained by physics-based simulations are not fully exploited by example-based or more recent datadriven methods neither, with most of them having focused on the modeling of static skin shapes by leveraging quality data. To address these limitations, we present a learningbased method for dynamic skin deformation. At the core of our work is a recurrent neural network that learns to predict the nonlinear, dynamics-dependent shape change over time from pre-existing mesh deformation sequence data. Our network also learns to predict the variation of skin dynamics across different individuals with varying body shapes. After training the network delivers realistic, high-quality skin dynamics that is specific to a person in a real-time course. We obtain results that significantly saves the computational time, while maintaining comparable prediction quality compared to state-of-the-art results.