Unpaired Image Enhancement Featuring Reinforcement-Learning-Controlled Image Editing Software
Unpaired Image Enhancement Featuring Reinforcement-Learning-Controlled Image Editing Software
复制标题
DOI：
10.1609/aaai.v34i07.6790
复制
发表时间：
2019-12
期刊：
影响因子：
--
通讯作者：
Satoshi Kosugi;T. Yamasaki
中科院分区：
文献类型：
--
作者：
Satoshi Kosugi;T. Yamasaki
本文解决了不成对图像增强问题，这是一项学习映射函数的任务，该映射函数在没有输入输出图像对的情况下将输入图像转换为增强图像。我们的方法基于生成对抗网络 (GAN)，但我们不是简单地使用神经网络生成图像，而是利用 Adobe® Photoshop® 等图像编辑软件增强图像，以获得以下三个好处：增强后的图像没有伪影、相同的增强可以应用于更大的图像，并且增强是可解释的。为了将图像编辑软件整合到 GAN 中，我们提出了一种强化学习框架，其中生成器充当选择软件参数的代理，并在欺骗鉴别器时获得奖励。我们的框架可以使用图像编辑软件中存在的高质量不可微滤波器，从而实现高性能图像增强。我们将所提出的方法应用于两个不成对的图像增强任务：照片增强和面部美化。我们的实验结果表明，与基于不配对学习的最先进方法的性能相比，所提出的方法取得了更好的性能。
This paper tackles unpaired image enhancement, a task of learning a mapping function which transforms input images into enhanced images in the absence of input-output image pairs. Our method is based on generative adversarial networks (GANs), but instead of simply generating images with a neural network, we enhance images utilizing image editing software such as Adobe® Photoshop® for the following three benefits: enhanced images have no artifacts, the same enhancement can be applied to larger images, and the enhancement is interpretable. To incorporate image editing software into a GAN, we propose a reinforcement learning framework where the generator works as the agent that selects the software's parameters and is rewarded when it fools the discriminator. Our framework can use high-quality non-differentiable filters present in image editing software, which enables image enhancement with high performance. We apply the proposed method to two unpaired image enhancement tasks: photo enhancement and face beautification. Our experimental results demonstrate that the proposed method achieves better performance, compared to the performances of the state-of-the-art methods based on unpaired learning.