BeautyGAN 閱讀筆記
@inproceedings{RN57,
author = {Li, Tingting and Qian, Ruihe and Dong, Chao and Liu, Si and Yan, Qiong and Zhu, Wenwu and Lin, Liang},
title = {Beautygan: Instance-level facial makeup transfer with deep generative adversarial network},
booktitle = {Proceedings of the 26th ACM international conference on Multimedia},
pages = {645-653},
type = {Conference Proceedings}
}
Contribution
這篇論文的主要contribution就是提出了 instance-level makeup transfer network BeautyGAN 和一個包含 makeup 和 non-makeup 圖片的數據集。
Important Points
- 採用衡量 face、lips、shadow 這三個地方的 pixel-level histogram loss;這裏還要先對人臉進行分割,然後使用histogram matching對齊。
- 提出了一個 3834 張圖片的 makeup 數據集;
Motivation
之前提出的 CycleGAN 只能實現domain level的轉換,makeup style 要求 instance level 的轉換,而且很難獲取到 pair of well-aligned face images with different makeup styles。