Path-Restore: Learning Network Path Selection for Image Restoration

• Published in: IEEE transactions on pattern analysis and machine intelligence 2022-10, Vol.44 (10), p.7078-7092
• Main Authors: Yu, Ke, Wang, Xintao, Dong, Chao, Tang, Xiaoou, Loy, Chen Change
• Format: Article
• Language: English
• Subjects: Artificial neural networks; Complexity theory; Computing costs; deep reinforcement learning; denoising; Distortion; dynamic network; Image restoration; Learning; Noise reduction; Reinforcement learning; Route selection; Task analysis; Training
• ISSN: 0162-8828; 2160-9292; 1939-3539
• DOI: 10.1109/TPAMI.2021.3096255

Very deep Convolutional Neural Networks (CNNs) have greatly improved the performance on various image restoration tasks. However, this comes at a price of increasing computational burden, hence limiting their practical usages. We observe that some corrupted image regions are inherently easier to restore than others since the distortion and content vary within an image. To leverage this, we propose Path-Restore, a multi-path CNN with a pathfinder that can dynamically select an appropriate route for each image region. We train the pathfinder using reinforcement learning with a difficulty-regulated reward. This reward is related to the performance, complexity and "the difficulty of restoring a region". A policy mask is further investigated to jointly process all the image regions. We conduct experiments on denoising and mixed restoration tasks. The results show that our method achieves comparable or superior performance to existing approaches with less computational cost. In particular, Path-Restore is effective for real-world denoising, where the noise distribution varies across different regions on a single image. Compared to the state-of-the-art RIDNet [1], our method achieves comparable performance and runs 2.7x faster on the realistic Darmstadt Noise Dataset [2]. Models and codes are available on the project page: https://www.mmlab-ntu.com/project/pathrestore/ .