Int J Performability Eng ›› 2019, Vol. 15 ›› Issue (3): 910-918.doi: 10.23940/ijpe.19.03.p20.910918

Previous Articles     Next Articles

Colorization for Anime Sketches with Cycle-Consistent Adversarial Network

Guanghua Zhang*, Mengnan Qu*,Yuhao Jin, and Qingpeng Song   

  1. School of Information Science and Engineering, Hebei University of Science and Technology, Shijiazhuang, 050018, China
  • Submitted on ; Revised on ;
  • Contact: xian_software@163.com
  • About author:Guanghua Zhang received his B.S. degree from Hebei Normal University, China, in 2002. He received his M.S. degree and Ph.D. from Xidian University, China, in 2005 and 2014 respectively. He is currently a vice professor in the College of Information Science and Engineering, Hebei University of Science and Technology, China. His main research interests include wireless security, trust management, and cooperative spectrum sensing.Mengnan Qu is an undergraduate student in the School of Information Science and Engineering, Hebei University of Science and Technology. His research interests include machine learning and digital image processing.Yuhao Jin is an undergraduate student in the School of Information Science and Engineering, Hebei University of Science and Technology. His research interests include machine learning and information security technology.Qingpeng Song is an undergraduate student in the School of Information Science and Engineering, Hebei University of Science and Technology. His research interests include machine learning.

Abstract: Coloring animation sketches has always been a complex and interesting task, but as the sketch is the first part of animation creation that neither presents gray value nor presents semantic information, the lack of real animation sketches is the biggest difficulty in current model training. It is also usually expensive to collect such data. In recent years, some methods based on generative adversarial networks (GANs) have achieved great success. They can generate colorized anime illustration on given sketches. Many existing sketch coloring tools are based on this supervised learning method, but the marking of data is particularly important for supervised learning, and much time is spent on the marking of data. To address these challenges, we propose a novel approach for unsupervised learning based on U-net and periodic consistent confrontation. Specifically, we combine the periodic consistent antagonism framework with the U-net structure and residual network, enabling us to robustly train a deep network to make the resulting images more natural and realistic. We also adopted some special data generation methods, so that our model can not only color anime sketches but also extract line drafts from colored pictures. By comparing the mainstream models of supervised learning, we show that the image processed by the proposed method can achieve a similar effect.

Key words: anime, sketches, colorization, cycle-consistency, u-net