TL;DR该论文提出了一种通过学习额外的代理来学习高保真人脸模型的新方法,结合全局和局部模型的精心设计网络,通过改进非线性 3D 可变形模型的学习目标和网络架构,使其在捕捉更高级别细节方面具有优越性,从而在 3D 人脸重建中实现了最先进的表现。
Abstract
Embedding 3d morphable basis functions into deep neural networks opens great
potential for models with better representation power. However, to faithfully
learn those models from an image collection, it requires