RestorNet: An efficient network for multiple degradation image restoration

Xiaofeng Wang, Honggang Chen*, Haosong Gou, Jie He, Zhengyong Wang, Xiaohai He, Linbo Qing, Ray E. Sheriff

*Corresponding author for this work

Research output: Contribution to journalArticle (journal)peer-review


Even though Single Degradation Image Restoration (SDIR) has made significant progress and achievedremarkable performance, Multiple Degradation Image
Restoration (MDIR) remains a long-term and arduous challenge to achieve the similar levels of success. To further improve the performance and efficiency of MDIR, we propose a novel MDIR method named RestorNet, which comprises an unsupervised degradation encoder for the learning of multi-scale degradation representations and a Multi-scale Degradation-assisted Restoration Module (MDRM) for image reconstruction. Our RestorNet aims to remove noise, rain, and haze in a unified network from the following three aspects. Firstly, to better distinguish among different degradations and learn the corruption information more accurately, we introduce a degradation-specific contrastive loss based on contrastive learning. Next, we develop a multi-scale degradation representation learning method to improve preservation of the spatial structure and distribution of inputs, and to extract multi-scale information to satisfy the diverse requirements of restoring different degraded images. Finally, to make a more reasonable use of degradation representation, we present a novel semi-guided strategy for effective feature transformation, where the multi-scale degradation representations are only incorporated into the MDRM encoder. For image denoising, deraining, and dehazing, by integrating the approaches above,
RestorNet not only outperforms the recent state-of-the-art MDIR algorithms with lower computational complexity, but also achieves impressive performance in SDIR. Extensive experiments demonstrate the effectiveness and superiority of our proposed method.
Original languageEnglish
Article number111116
Number of pages23
JournalKnowledge-Based Systems
Early online date28 Oct 2023
Publication statusPublished - 20 Dec 2023


  • Multiple Degradation Image Restoration
  • Image Denoising, Deraining and Dehazing
  • Degradation-specific Contrastive Learning
  • Multi-scale Degradation Representations
  • Semi-guided Strategy
  • Image denoising, deraining and dehazing
  • Semi-guided strategy
  • Multi-scale degradation representations
  • Multiple degradation image restoration
  • Degradation-specific contrastive learning


Dive into the research topics of 'RestorNet: An efficient network for multiple degradation image restoration'. Together they form a unique fingerprint.

Cite this