Single-image reflection removal : from computational imaging to deep learning

Reflection removal aims at enhancing the visibility of the background scene while removing the reflections for images taken through the transparent glass. Though it is of broad application to various computer vision tasks, it is very challenging due to its ill-posed nature and additional priors are...

Full description

Saved in:
Bibliographic Details
Main Author: Wan, Renjie
Other Authors: Alex C. Kot
Format: Theses and Dissertations
Language:English
Published: 2019
Subjects:
Online Access:https://hdl.handle.net/10356/82986
http://hdl.handle.net/10220/47556
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-82986
record_format dspace
spelling sg-ntu-dr.10356-829862020-11-01T04:55:15Z Single-image reflection removal : from computational imaging to deep learning Wan, Renjie Alex C. Kot Interdisciplinary Graduate School (IGS) DRNTU::Engineering::Electrical and electronic engineering Reflection removal aims at enhancing the visibility of the background scene while removing the reflections for images taken through the transparent glass. Though it is of broad application to various computer vision tasks, it is very challenging due to its ill-posed nature and additional priors are needed to make this problem tractable. Traditional reflection removal methods solve this problem by making use of different heuristic observations or assumptions. These assumptions are seldom satisfied in practical scenarios. In this thesis, we generalize the assumptions for the reflection removal problems by using different information or imposing new constraints. We first propose a method by exploring the blur inconsistency between the background and reflections. Then, we introduce the first benchmark dataset in this area and analyze limitations of existing methods based on this dataset. In the third work, we address this problem by using the sparsity prior and non-local image prior from the external source. Then, with the observation that most reflections only cover a part of the whole image, we propose a method to automatically detect the regions with and without reflections and process them in a heterogeneous manner. At last, we introduce a data-driven method by using the concurrent deep learning framework. Our methods have been evaluated by using the benchmark dataset proposed in our second work. These evaluations cover a diversity of common scenarios in our daily life; hence the experiments prove that our approaches are valid for a broad class of practical scenarios. The main contributions of this thesis are three folds: We thoroughly study the reflection properties observed in our daily scenarios; we propose the first benchmark evaluation dataset in this area and use it to analyze the limitations of existing methods; we propose various approaches to solve this problem from different angles. The efforts and achievements in this thesis promote the practical capabilities of reflection removal techniques and provide fundamental support for future researches. Doctor of Philosophy 2019-01-24T08:43:36Z 2019-12-06T15:09:40Z 2019-01-24T08:43:36Z 2019-12-06T15:09:40Z 2018 Thesis Wan, R. (2018). Single-image reflection removal : from computational imaging to deep learning. Doctoral thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/82986 http://hdl.handle.net/10220/47556 10.32657/10220/47556 en 161 p. application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic DRNTU::Engineering::Electrical and electronic engineering
spellingShingle DRNTU::Engineering::Electrical and electronic engineering
Wan, Renjie
Single-image reflection removal : from computational imaging to deep learning
description Reflection removal aims at enhancing the visibility of the background scene while removing the reflections for images taken through the transparent glass. Though it is of broad application to various computer vision tasks, it is very challenging due to its ill-posed nature and additional priors are needed to make this problem tractable. Traditional reflection removal methods solve this problem by making use of different heuristic observations or assumptions. These assumptions are seldom satisfied in practical scenarios. In this thesis, we generalize the assumptions for the reflection removal problems by using different information or imposing new constraints. We first propose a method by exploring the blur inconsistency between the background and reflections. Then, we introduce the first benchmark dataset in this area and analyze limitations of existing methods based on this dataset. In the third work, we address this problem by using the sparsity prior and non-local image prior from the external source. Then, with the observation that most reflections only cover a part of the whole image, we propose a method to automatically detect the regions with and without reflections and process them in a heterogeneous manner. At last, we introduce a data-driven method by using the concurrent deep learning framework. Our methods have been evaluated by using the benchmark dataset proposed in our second work. These evaluations cover a diversity of common scenarios in our daily life; hence the experiments prove that our approaches are valid for a broad class of practical scenarios. The main contributions of this thesis are three folds: We thoroughly study the reflection properties observed in our daily scenarios; we propose the first benchmark evaluation dataset in this area and use it to analyze the limitations of existing methods; we propose various approaches to solve this problem from different angles. The efforts and achievements in this thesis promote the practical capabilities of reflection removal techniques and provide fundamental support for future researches.
author2 Alex C. Kot
author_facet Alex C. Kot
Wan, Renjie
format Theses and Dissertations
author Wan, Renjie
author_sort Wan, Renjie
title Single-image reflection removal : from computational imaging to deep learning
title_short Single-image reflection removal : from computational imaging to deep learning
title_full Single-image reflection removal : from computational imaging to deep learning
title_fullStr Single-image reflection removal : from computational imaging to deep learning
title_full_unstemmed Single-image reflection removal : from computational imaging to deep learning
title_sort single-image reflection removal : from computational imaging to deep learning
publishDate 2019
url https://hdl.handle.net/10356/82986
http://hdl.handle.net/10220/47556
_version_ 1683493869293928448