# 论文阅读 [TPAMI-2022] GarNet++: Improving Fast and Accurate Static 3D Cloth Draping by Curvature Loss

论文阅读 [TPAMI-2022] GarNet++: Improving Fast and Accurate Static 3D Cloth Draping by Curvature Loss

论文搜索(studyai.com)

搜索论文: GarNet++: Improving Fast and Accurate Static 3D Cloth Draping by Curvature Loss

搜索论文: http://www.studyai.com/search/whole-site/?q=GarNet++:+Improving+Fast+and+Accurate+Static+3D+Cloth+Draping+by+Curvature+Loss

关键字(Keywords)

Clothing; Three-dimensional displays; Shape; Computational modeling; Solid modeling; Biological system modeling; Deformable models; 3D point cloud; garment draping; mesh convolution; physics-based simulation

机器视觉; 运筹与优化

虚拟试穿; 损失函数; 三维点云

摘要(Abstract)

In this paper, we tackle the problem of static 3D cloth draping on virtual human bodies.

在本文中，我们解决了虚拟人体上静态3D布料覆盖的问题。.

We introduce a two-stream deep network model that produces a visually plausible draping of a template cloth on virtual 3D bodies by extracting features from both the body and garment shapes.

我们介绍了一个两流深度网络模型，该模型通过从身体和服装形状中提取特征，在虚拟3D身体上生成视觉上合理的模板布料覆盖。.

Our network learns to mimic a physics-based simulation (PBS) method while requiring two orders of magnitude less computation time.

我们的网络学习模仿基于物理的模拟（PBS）方法，同时需要两个数量级的计算时间。.

To train the network, we introduce loss terms inspired by PBS to produce plausible results and make the model collision-aware.

为了训练网络，我们引入了受PBS启发的损失项，以产生合理的结果，并使模型能够感知碰撞。.

To increase the details of the draped garment, we introduce two loss functions that penalize the difference between the curvature of the predicted cloth and PBS.

为了增加褶皱服装的细节，我们引入了两个损失函数来惩罚预测布料和PBS曲率之间的差异。.

Particularly, we study the impact of mean curvature normal and a novel detail-preserving loss both qualitatively and quantitatively.

特别地，我们从定性和定量两方面研究了平均曲率法向和一种新的细节保持损失的影响。.

Our new curvature loss computes the local covariance matrices of the 3D points, and compares the Rayleigh quotients of the prediction and PBS.

我们的新曲率损失计算3D点的局部协方差矩阵，并比较预测和PBS的瑞利商。.

This leads to more details while performing favorably or comparably against the loss that considers mean curvature normal vectors in the 3D triangulated meshes.

这会在考虑3D三角网格中的平均曲率法向量的损失时，产生更多细节，同时表现良好或类似。.

We validate our framework on four garment types for various body shapes and poses.

我们在四种不同体型和姿势的服装类型上验证了我们的框架。.

Finally, we achieve superior performance against a recently proposed data-driven method…

最后，与最近提出的数据驱动方法相比，我们获得了更高的性能。。.

作者(Authors)

[‘Erhan Gundogdu’, ‘Victor Constantin’, ‘Shaifali Parashar’, ‘Amrollah Seifoddini’, ‘Minh Dang’, ‘Mathieu Salzmann’, ‘Pascal Fua’]