Real-Time Patch-Based Stylization of Portraits Using Generative Adversarial Network
Loading...
Date
2019
Journal Title
Journal ISSN
Volume Title
Publisher
The Eurographics Association
Abstract
We present a learning-based style transfer algorithm for human portraits which significantly outperforms current state-of-the-art in computational overhead while still maintaining comparable visual quality. We show how to design a conditional generative adversarial network capable to reproduce the output of Fišer et al.'s patch-based method [FJS*17] that is slow to compute but can deliver state-of-the-art visual quality. Since the resulting end-to-end network can be evaluated quickly on current consumer GPUs, our solution enables first real-time high-quality style transfer to facial videos that runs at interactive frame rates. Moreover, in cases when the original algorithmic approach of Fišer et al. fails our network can provide a more visually pleasing result thanks to generalization. We demonstrate the practical utility of our approach on a variety of different styles and target subjects.
Description
@inproceedings{10.2312:exp.20191074,
booktitle = {ACM/EG Expressive Symposium},
editor = {Kaplan, Craig S. and Forbes, Angus and DiVerdi, Stephen},
title = {{Real-Time Patch-Based Stylization of Portraits Using Generative Adversarial Network}},
author = {Futschik, David and Chai, Menglei and Cao, Chen and Ma, Chongyang and Stoliar, Aleksei and Korolev, Sergey and Tulyakov, Sergey and Kučera, Michal and Sýkora, Daniel},
year = {2019},
publisher = {The Eurographics Association},
ISBN = {978-3-03868-078-9},
DOI = {10.2312/exp.20191074}
}