Cited 0 time in webofscience Cited 0 time in scopus

Real-World Blur Dataset for Learning and Benchmarking Deblurring Algorithms

Title
Real-World Blur Dataset for Learning and Benchmarking Deblurring Algorithms
Authors
Rim, Jaesung
DGIST Authors
Rim, Jaesung; Chwa, Hoon Sung; Cho, Sunghyun
Advisor(s)
좌훈승
Co-Advisor(s)
Cho, Sunghyun
Issue Date
2020
Available Date
2020-08-06
Degree Date
2020/08
Type
Thesis
Description
Computational Photography, Deblurring, Low-level Vision, Datasets and Evaluation
Abstract
Numerous learning-based approaches to single image deblurring for camera and object motion blurs have recently been proposed. To generalize such approaches to real-world blurs, large datasets of real blurred images and their ground truth sharp images are essential. However, there are still no such datasets, thus all the existing approaches resort to synthetic ones, which leads to the failure of deblurring real-world images. In this work, we present a large-scale dataset of real-world blurred images and their corresponding sharp images captured in low-light environments for learning and benchmarking single image deblurring methods. To collect our dataset, we build an image acquisition system to simultaneously capture a geometrically aligned pair of blurred and sharp images, and develop a post-processing method to further align images geometrically and photometrically. We analyze the effect of our post-processing step, and the performance of existing learning-based deblurring methods. Our analysis shows that our dataset significantly improves deblurring quality for real-world low-light images.
Table Of Contents
1. Introduction 1 2. Related Work 2 3. Image Acquisition System and Process 3 3.1 Image Acquisition System 3 3.2 Image Acquisition Process 4 4. Post-Processing 5 4.1 Downsampling & Denoising 6 4.2 Geometric Alignment 6 4.3 Photometric Alignment 8 5. Experiments 8 5.1 Analysis of RealBlur Dataset 9 5.2 Benchmark 12 6. Conclusion 19 7. Appendix 20 8. References 24 9. 요약문 28
URI
http://dgist.dcollection.net/common/orgView/200000333386
http://hdl.handle.net/20.500.11750/12177
DOI
https://doi.org/10.22677/thesis.200000333386
Degree
Master
Department
Department of Information and Communication Engineering
University
DGIST
Related Researcher
  • Author Chwa, Hoon Sung Real-Time Computing Lab
  • Research Interests Real-Time Systems; Real-Time AI Services; Cyber-Physical Systems; Mobile Systems
Files:
Collection:
Department of Information and Communication EngineeringThesesMaster


qrcode mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE