Single Image Reflection Removal Based on GAN with Gradient Constraint

Ryo Abiko, Masaaki Ikehara

研究成果: Conference contribution

抄録

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 network. 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 minimize 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.

本文言語English
ホスト出版物のタイトルPattern Recognition - 5th Asian Conference, ACPR 2019, Revised Selected Papers
編集者Shivakumara Palaiahnakote, Gabriella Sanniti di Baja, Liang Wang, Wei Qi Yan
出版社Springer
ページ609-624
ページ数16
ISBN(印刷版)9783030414030
DOI
出版ステータスPublished - 2020
イベント5th Asian Conference on Pattern Recognition, ACPR 2019 - Auckland, New Zealand
継続期間: 2019 11 262019 11 29

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
12046 LNCS
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Conference

Conference5th Asian Conference on Pattern Recognition, ACPR 2019
国/地域New Zealand
CityAuckland
Period19/11/2619/11/29

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「Single Image Reflection Removal Based on GAN with Gradient Constraint」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル