A New Makeup Transfer with Super-resolution.
Yeqing Ren, Youqiang Sun, Di Wu, Zhihua Cui, Alex Kai Qin · Swinburne figshare (Swinburne University of Technology) · 2019
With the wide use of beauty camera, the makeup transfer for old photo has a certain application value. In this paper, we propose a new makeup transfer model considering a specific scene that is blurry target image. We use SRWGAN-GP to reconstruct the blurred target image to make it clear and then create face makeup upon it with example image as makeup style. And we take the beauty in Baidu AI as the evaluation standard of makeup. Experimental results demonstrate the effectiveness of our method, and it greatly improves the beauty of the blurry target image. What’s more, our method has certain application value, it provides an effective way to transfer makeup for old photo, and the result image is very clear, which is more valuable for users than the blurred photo using original makeup transfer method.