Distort-and-Recover: Color Enhancement Using Deep Reinforcement Learning
Authors:
- Jongchan ParkJPJongchan Park
- Joon-Young LeeJLJoon-Young Lee
- Donggeun YooDYDonggeun Yoo
- In So KweonIKIn So Kweon
- 1 June 2018
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 5928-5936
Abstract
Learning-based color enhancement approaches typically learn to map from input images to retouched images. Most of existing methods require expensive pairs of input-retouched images or produce results in a non-interpretable way. In this paper, we present a deep reinforcement learning (DRL) based method for color enhancement to explicitly model the step-wise nature of human retouching process. We cast a color enhancement process as a Markov Decision Process where actions are defined as global color adjustment operations. Then we train our agent to learn the optimal global enhancement sequence of the actions. In addition, we present a 'distort-and-recover' training scheme which only requires high-quality reference images for training instead of input and retouched image pairs. Given high-quality reference images, we distort the images' color distribution and form distorted-reference image pairs for training. Through extensive experiments, we show that our method produces decent enhancement results and our DRL approach is more suitable for the 'distort-and-recover' training scheme than previous supervised approaches. Supplementary material and code are available at https://sites.google.com/view/distort-and-recover/.Keywords
All Related Versions
This publication has 13 references indexed in Scilit:
- Photo Aesthetics Ranking Network with Attributes and Content Adaptation2016
- Automatic Photo Adjustment Using Deep Neural NetworksACM Transactions on Graphics, 2016
- Deep Multi-patch Aggregation Network for Image Style, Aesthetics, and Quality Estimation2015
- Multi-scale pyramid pooling for deep convolutional representation2015
- ImageNet Large Scale Visual Recognition ChallengeInternational Journal of Computer Vision, 2015
- Human-level control through deep reinforcement learningNature, 2015
- RAPID2014
- AutoStyle: Automatic Style Transfer from Image Collections to Users' ImagesComputer Graphics Forum, 2014
- AVA: A large-scale database for aesthetic visual analysis2012
- Color transfer between imagesIEEE Computer Graphics and Applications, 2001