3D Face From X: Learning Face Shape from Diverse Sources
This addresses the challenge of limited and expensive 3D face data for computer vision researchers, though it appears incremental by combining existing sources.
The paper tackles the problem of learning a 3D face parametric model and reconstruction from diverse sources, such as scanned data, in-the-wild images, and RGB-D images, and shows that using more sources leads to a more powerful model.
We present a novel method to jointly learn a 3D face parametric model and 3D face reconstruction from diverse sources. Previous methods usually learn 3D face modeling from one kind of source, such as scanned data or in-the-wild images. Although 3D scanned data contain accurate geometric information of face shapes, the capture system is expensive and such datasets usually contain a small number of subjects. On the other hand, in-the-wild face images are easily obtained and there are a large number of facial images. However, facial images do not contain explicit geometric information. In this paper, we propose a method to learn a unified face model from diverse sources. Besides scanned face data and face images, we also utilize a large number of RGB-D images captured with an iPhone X to bridge the gap between the two sources. Experimental results demonstrate that with training data from more sources, we can learn a more powerful face model.