Learning Body Shape and Pose from Dense Correspondences
Learning Body Shape and Pose from Dense Correspondences
复制标题
DOI：
10.2312/egs.20201012
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
Y. Yoshiyasu;L. Gamez
中科院分区：
文献类型：
--
作者：
Y. Yoshiyasu;L. Gamez
在本文中，我们解决了从 2D 图像数据集学习 3D 人体姿势和身体形状的问题，而无需使用 3D 数据集（身体形状和姿势）。这个想法是利用图像点和身体表面之间的密集对应关系，这些对应关系可以在野外 2D 图像上进行注释，并从中提取和聚合 3D 信息。为此，我们提出了一种名为“变形并学习”的训练策略，其中我们交替使用可变形表面配准和深度卷积神经网络 (ConvNet) 的训练。与以前的方法不同，我们的方法不需要来自运动捕捉 (MoCap) 系统的 3D 姿势注释或人工干预来验证 3D 姿势注释。
In this paper, we address the problem of learning 3D human pose and body shape from 2D image dataset, without having to use 3D dataset (body shape and pose). The idea is to use dense correspondences between image points and a body surface, which can be annotated on in-the wild 2D images, and extract and aggregate 3D information from them. To do so, we propose a training strategy called ``deform-and-learn" where we alternate deformable surface registration and training of deep convolutional neural networks (ConvNets). Unlike previous approaches, our method does not require 3D pose annotations from a motion capture (MoCap) system or human intervention to validate 3D pose annotations.