Single image reflection removal beyond linearity

Due to the lack of paired data, the training of image reflection removal relies heavily on synthesizing reflection images. However, existing methods model reflection as a linear combination model, which cannot fully simulate the real-world scenarios. In this paper, we inject non-linearity into refle...

Full description

Saved in:
Bibliographic Details
Main Authors: WEN, Qiang, TAN, Yinjie, QIN, Jing, LIU, Wenxi, HAN, Guoqiang, HE, Shengfeng
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2019
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/8437
https://ink.library.smu.edu.sg/context/sis_research/article/9440/viewcontent/Wen_Single_Image_Reflection_Removal_Beyond_Linearity_CVPR_2019_paper.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
Description
Summary:Due to the lack of paired data, the training of image reflection removal relies heavily on synthesizing reflection images. However, existing methods model reflection as a linear combination model, which cannot fully simulate the real-world scenarios. In this paper, we inject non-linearity into reflection removal from two aspects. First, instead of synthesizing reflection with a fixed combination factor or kernel, we propose to synthesize reflection images by predicting a non-linear alpha blending mask. This enables a free combination of different blurry kernels, leading to a controllable and diverse reflection synthesis. Second, we design a cascaded network for reflection removal with three tasks: predicting the transmission layer, reflection layer, and the non-linear alpha blending mask. The former two tasks are the fundamental outputs, while the latter one being the side output of the network. This side output, on the other hand, making the training a closed loop, so that the separated transmission and reflection layers can be recombined together for training with a reconstruction loss. Extensive quantitative and qualitative experiments demonstrate the proposed synthesis and removal approaches outperforms state-of-the-art methods on two standard benchmarks, as well as in real-world scenarios.