跳到正文
返回论文列表
cs.GR提交于 已译

基于姿态的动画角色神经服装

Pose-based Neural Clothing for Animated Characters

Julian N. Heidenreich · Vinicius C. Azevedo · Jakob Buhmann · Lento Manickathan · Arnold Moon · Paul Kanyuk · et al.

中文摘要

传统动画角色的服装模拟因仿真缓慢和参数调优繁琐,成为美术师耗时瓶颈。数据驱动方法能从模拟数据集学习高质量变形,轻量网络设计可实现实时推理,显著改善工作流,让美术师交互式摆姿势并即时看到高质量褶皱。然而,现有神经服装方法在真实人体模型(如SMPL)数据集上表现良好,但在动画角色数据集上训练时难以泛化到未见姿态。本文旨在设计神经网络和训练策略,以实时更好地泛化到未见姿态,支持交互式摆姿势应用。聚焦于生产导向的单件服装准静态设置,利用平滑、表达性强的潜在空间,先训练变分自编码器(VAE),再逐步训练VAE编码器从重建过渡到姿态驱动预测。网络以低维姿态参数和角色身体表面位移为条件。引入新颖的局部梯度直方图损失,缓解重建损失的过度平滑效应。在生产级资产和公开数据集上的广泛评估表明,该方法能实时生成锐利逼真的褶皱、平滑的大尺度悬垂和时间连贯性,为交互式摆姿势工作流提供稳健且易于集成的解决方案。

关键要点

  1. 01问题:现有神经服装方法在动画角色数据集上难以泛化到未见姿态。
  2. 02方法:先训练VAE获得平滑潜在空间,再逐步训练编码器从重建转向姿态驱动预测。
  3. 03方法:网络以低维姿态参数和身体表面位移为条件,并引入局部梯度直方图损失。
  4. 04结果:在生产级资产和公开数据集上实时生成锐利褶皱、平滑悬垂和时间连贯性。

解读

尚无解读。

原始英文摘要

arXiv:2611.18253v1 Announce Type: cross Abstract: Traditional garment simulation for animated characters is a time-consuming bottleneck for artists due to slow simulations and tedious parameter tunings.Recent data-driven methods offer a powerful alternative, capable of learning from a dataset of simulated data to generate high-quality deformations. A lightweight network design allows data-driven systems to achieve a real-time inference speed. This significantly improves the artist workflow, as they can interactively pose a character and see high-quality folds and wrinkles at animation time, thereby allowing to make creative decisions ahead of time. However, existing neural clothing methods have presented appealing results on datasets based on realistic human models (\eg SMPL), but when trained on animated character datasets, they struggle to generalize to unseen poses. In this paper, we aim at designing neural networks and training strategies to generalize better to unseen poses in real-time to enable the interactive posing applications. We focus on a production-oriented setting where the model is trained for a single garment in a quasi-static setup. Technically, we leverage a smooth, expressive latent space by first training a Variational Auto-Encoder (VAE). We then progressively train the VAE encoder to transition from reconstruction to pose-driven prediction. To guide this prediction, we condition the network on both low-dimensional pose parameters and character body surface displacements. Finally, we introduce a novel local gradient histogram loss that mitigates the over-smoothing effect of reconstruction losses. Extensive evaluations on production-level assets and public datasets demonstrate that our method produces sharp, realistic wrinkles, smooth large-scale drape, and temporal coherence in real-time, offering a robust and easily integrable solution for interactive posing workflows.

同方向论文 · cs.GR

查看全部 →