4.7 Article

GAN-Based Multi-Style Photo Cartoonization

期刊

出版社

IEEE COMPUTER SOC
DOI: 10.1109/TVCG.2021.3067201

关键词

Training; Generative adversarial networks; Semantics; Image edge detection; Training data; Generators; Computer architecture; Style transfer; cartoon styles; multi-style transfer; generative adversarial network

资金

  1. Natural Science Foundation of China [61725204]

向作者/读者索取更多资源

In this article, a novel multi-style generative adversarial network (GAN) architecture called MS-CartoonGAN is proposed, which can transform photos into multiple cartoon styles. The shared network architecture exploits the common characteristics of cartoon styles, achieving better cartoonization and being more efficient than single-style cartoonization methods.
Cartoon is a common form of art in our daily life and automatic generation of cartoon images from photos is highly desirable. However, state-of-the-art single-style methods can only generate one style of cartoon images from photos and existing multi-style image style transfer methods still struggle to produce high-quality cartoon images due to their highly simplified and abstract nature. In this article, we propose a novel multi-style generative adversarial network (GAN) architecture, called MS-CartoonGAN, which can transform photos into multiple cartoon styles. MS-CartoonGAN uses only unpaired photos and cartoon images of multiple styles for training. To achieve this, we propose to use (1) a hierarchical semantic loss with sparse regularization to retain semantic content and recover flat shading in different abstract levels, (2) a new edge-promoting adversarial loss for producing fine edges, and (3) a style loss to enhance the difference between output cartoon styles and make training process more stable. We also develop a multi-domain architecture, where the generator consists of a shared encoder and multiple decoders for different cartoon styles, along with multiple discriminators for individual styles. By observing that cartoon images drawn by different artists have their unique styles while sharing some common characteristics, our shared network architecture exploits the common characteristics of cartoon styles, achieving better cartoonization and being more efficient than single-style cartoonization. We show that our multi-domain architecture can theoretically guarantee to output desired multiple cartoon styles. Through extensive experiments including a user study, we demonstrate the superiority of the proposed method, outperforming state-of-the-art single-style and multi-style image style transfer methods.

作者

我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。

评论

主要评分

4.7
评分不足

次要评分

新颖性
-
重要性
-
科学严谨性
-
评价这篇论文

推荐

暂无数据
暂无数据