Two-step domain adaptation for underwater image enhancement
作者：
Highlights：
• Inspired by transfer learning, we migrate in-air image dehazing to underwater image enhancement.
• We propose a novel two-step domain adaptation framework for underwater image enhancement, which realizes cross-domain adaptation from the air domain to the underwater domain.
• Our method is trained on real-world underwater images without utilizing underwater images synthesized with in-air images, which eliminates the dependence on underwater paired data.
摘要
•Inspired by transfer learning, we migrate in-air image dehazing to underwater image enhancement.•We propose a novel two-step domain adaptation framework for underwater image enhancement, which realizes cross-domain adaptation from the air domain to the underwater domain.•Our method is trained on real-world underwater images without utilizing underwater images synthesized with in-air images, which eliminates the dependence on underwater paired data.
论文关键词：Underwater image enhancement,Transfer learning,Domain adaptation,Cycle-consistent adversarial network
论文评审过程：Received 12 March 2021, Revised 4 July 2021, Accepted 11 September 2021, Available online 12 September 2021, Version of Record 22 September 2021.
论文官网地址：https://doi.org/10.1016/j.patcog.2021.108324
