Single Image Reflection Removal Based on GAN with Gradient Constraint

Ryo Abiko, Masaaki Ikehara

Research output: Contribution to journalArticlepeer-review

12 Citations (Scopus)


When we take a picture through glass windows, the photographs are often degraded by undesired reflections. To separate reflection layer and background layer is an important problem for enhancing image quality. However, single-image reflection removal is a challenging process because of the ill-posed nature of the problem. In this paper, we propose a single-image reflection removal method based on generative adversarial networks. Our network is an end-To-end trained network with four types of losses. It includes pixel loss, feature loss, adversarial loss, and gradient constraint loss. We propose a novel gradient constraint loss in order to separate the background layer and the reflection layer clearly. Gradient constraint loss is applied in a gradient domain and it minimizes the correlation between the background and reflection layer. Owing to the novel loss and our new synthetic dataset, our reflection removal method outperforms state-of-The-Art methods in PSNR and SSIM, especially in real world images.

Original languageEnglish
Article number8868089
Pages (from-to)148790-148799
Number of pages10
JournalIEEE Access
Publication statusPublished - 2019


  • Image restoration
  • deep learning
  • generative adversarial network
  • image separation
  • reflection removal

ASJC Scopus subject areas

  • Computer Science(all)
  • Materials Science(all)
  • Engineering(all)
  • Electrical and Electronic Engineering


Dive into the research topics of 'Single Image Reflection Removal Based on GAN with Gradient Constraint'. Together they form a unique fingerprint.

Cite this