Learning to simultaneously handle face alignment of arbitrary views, e.g.
frontal and profile views, appears to be more challenging than we thought. The
difficulties lay in i) accommodating the complex appearance-shape relations
exhibited in different views, and ii) encompassing the va
本文提出了一种新的面部对齐框架,称为 3D Dense Face Alignment (3DDFA),其中使用级联卷积神经网络将密集 3D 可变形模型(3DMM)配合到图像中。本文还利用 3D 信息合成正面视图和侧面视图的面部图像以提供丰富的样本进行训练。在具有挑战性的 AFLW 数据库上的实验表明,所提出的方法比现有技术取得了显着的进步。