aim 2020 challenge on learned image signal processing pipeline

• DOI: 10.48550/arxiv.2011.04994
• Publication Date: 2020-01-01
• Authors: Yeskendir Koishek...; Jun Jiang; Bingxin Hou; Chengqi Li; Tian Liang; Jinwei Gu; Xin Liu; Kai Li; Byung-Hoon Kim; Yu Zhu; Haolin Wang; Zhilu Zhang; Umapada Pal; Ruimao Zhang; Xiaohong Liu; Puneesh Deora; Ming Liu; Zhihong Pan; Zhenyu Guo; Linhui Dai; Radu Timofte; Wangmeng Zuo; Jong Chul Ye; Pengliang Tan; Chenghua Li; Zhanglin Peng; Hwechul Cho Ye; Yuichi Ito; Xueying Hu; Bhavya Vasudeva; Magauiya Zhussip; JaeHyun Baek; Jiawei Zhang; Sijie Ren; Jun Chen; Baopu Li; Joonyoung Song; Andrey Ignatov; Cong Leng
• Keywords: FOS: Computer and information sciences; Computer Vision and Pattern Recognition (cs.CV); Image and Video Processing (eess.IV); Computer Science - Computer Vision and Pattern Recognition; 0202 electrical engineering, electronic engineering, information engineering; FOS: Electrical engineering, electronic engineering, information engineering; 02 engineering and technology; Electrical Engineering and Systems Science - Image and Video Processing

This paper reviews the second AIM learned ISP challenge and provides the description of the proposed solutions and results. The participating teams were solving a real-world RAW-to-RGB mapping problem, where to goal was to map the original low-quality RAW images captured by the Huawei P20 device to the same photos obtained with the Canon 5D DSLR camera. The considered task embraced a number of complex computer vision subtasks, such as image demosaicing, denoising, white balancing, color and contrast correction, demoireing, etc. The target metric used in this challenge combined fidelity scores (PSNR and SSIM) with solutions' perceptual results measured in a user study. The proposed solutions significantly improved the baseline results, defining the state-of-the-art for practical image signal processing pipeline modeling.<br/>Published in ECCV 2020 Workshops (Advances in Image Manipulation), https://data.vision.ee.ethz.ch/cvl/aim20/<br/>