arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.01247cs.CV
收藏

H4D:通过学习神经组成表示的人体4D建模

H4D: Human 4D Modeling by Learning Neural Compositional Representation

Fudan University(复旦大学);Google(谷歌)

Boyan Jiang, Yinda Zhang, Xingkui Wei, Xiangyang Xue, Yanwei Fu

摘要:

尽管基于深度学习的3D重建取得了令人印象深刻的结果,但直接学习以详细几何形状对4D人体捕获进行建模的技术研究较少。这项工作提出了一种新的框架,可以有效地学习一个紧凑的和组成的表示动态人体利用人体之前,从广泛使用的SMPL参数模型。特别是,我们的表示,命名为H4D,代表了一个动态的三维人类在一个时间跨度与SMPL参数的形状和初始姿态,和潜在的代码编码的运动和辅助信息。提出了一种简单而有效的线性运动模型,以提供粗略和正则化的运动估计,然后利用辅助代码中编码的残差对姿态和几何细节进行逐帧补偿。在技术上,我们引入了新的基于GRU的体系结构,以促进学习和提高表示能力。大量的实验表明,我们的方法不仅是有效的恢复动态人体准确的运动和详细的几何形状,但也适合各种4D人体相关的任务,包括运动重定向,运动完成和未来的预测。请查看项目页面的视频和代码:https://boyanjiang.github.io/H4D/。

英文摘要:

Despite the impressive results achieved by deep learning based 3D reconstruction, the techniques of directly learning to model 4D human captures with detailed geometry have been less studied. This work presents a novel framework that can effectively learn a compact and compositional representation for dynamic human by exploiting the human body prior from the widely used SMPL parametric model. Particularly, our representation, named H4D, represents a dynamic 3D human over a temporal span with the SMPL parameters of shape and initial pose, and latent codes encoding motion and auxiliary information. A simple yet effective linear motion model is proposed to provide a rough and regularized motion estimation, followed by per-frame compensation for pose and geometry details with the residual encoded in the auxiliary code. Technically, we introduce novel GRU-based architectures to facilitate learning and improve the representation capability. Extensive experiments demonstrate our method is not only efficacy in recovering dynamic human with accurate motion and detailed geometry, but also amenable to various 4D human related tasks, including motion retargeting, motion completion and future prediction. Please check out the project page for video and code: https://boyanjiang.github.io/H4D/.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑