Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera

Despite CNN-based deblur models have shown their superiority when solving motion blurs, restoring a photorealistic image from severe motion blur remains an ill-posed problem due to the loss of temporal information and textures. Event cameras such as Dynamic and Active-pixel Vision Sensor (DAVIS) can...

Full description

Saved in:

Bibliographic Details
Main Authors:	Limeng Zhang, Hongguang Zhang, Jihua Chen, Lei Wang
Format:	article
Language:	EN
Published:	IEEE 2020
Subjects:	Event-based vision high speed image deblurring real-time Electrical engineering. Electronics. Nuclear engineering TK1-9971
Online Access:	https://doaj.org/article/a23fb01e75674ee2b0894170b13cb055
Tags:	Add Tag No Tags, Be the first to tag this record!

id	oai:doaj.org-article:a23fb01e75674ee2b0894170b13cb055
record_format	dspace
spelling	oai:doaj.org-article:a23fb01e75674ee2b0894170b13cb0552021-11-19T00:03:37ZHybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera2169-353610.1109/ACCESS.2020.3015759https://doaj.org/article/a23fb01e75674ee2b0894170b13cb0552020-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9165110/https://doaj.org/toc/2169-3536Despite CNN-based deblur models have shown their superiority when solving motion blurs, restoring a photorealistic image from severe motion blur remains an ill-posed problem due to the loss of temporal information and textures. Event cameras such as Dynamic and Active-pixel Vision Sensor (DAVIS) can simultaneously produce gray-scale Active Pixel Sensor (APS) frames and events, which can capture fast motions as events of very high temporal resolution, <italic>i. e.</italic>, <inline-formula> <tex-math notation="LaTeX">$1~\mu s$ </tex-math></inline-formula>, can provide extra information for blurry APS frames. Due to the natural noise and sparsity of events, we employ a recurrent encoder-decoder architecture to generate dense recurrent event representations, which encode the overall historical information. We concatenate the original blurry image with the event representation as our hybrid input, from which the network learns to restore the sharp output. We conduct extensive experiments on GoPro dataset and a real event blurry dataset captured by DAVIS240C. Our experimental results on both synthetic and real images demonstrate state-of-the-art performance for <inline-formula> <tex-math notation="LaTeX">$1280\times 720 $ </tex-math></inline-formula> images at 30 fps.Limeng ZhangHongguang ZhangJihua ChenLei WangIEEEarticleEvent-based visionhigh speedimage deblurringreal-timeElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 8, Pp 148075-148083 (2020)
institution	DOAJ
collection	DOAJ
language	EN
topic	Event-based vision high speed image deblurring real-time Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Event-based vision high speed image deblurring real-time Electrical engineering. Electronics. Nuclear engineering TK1-9971 Limeng Zhang Hongguang Zhang Jihua Chen Lei Wang Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera
description	Despite CNN-based deblur models have shown their superiority when solving motion blurs, restoring a photorealistic image from severe motion blur remains an ill-posed problem due to the loss of temporal information and textures. Event cameras such as Dynamic and Active-pixel Vision Sensor (DAVIS) can simultaneously produce gray-scale Active Pixel Sensor (APS) frames and events, which can capture fast motions as events of very high temporal resolution, <italic>i. e.</italic>, <inline-formula> <tex-math notation="LaTeX">$1~\mu s$ </tex-math></inline-formula>, can provide extra information for blurry APS frames. Due to the natural noise and sparsity of events, we employ a recurrent encoder-decoder architecture to generate dense recurrent event representations, which encode the overall historical information. We concatenate the original blurry image with the event representation as our hybrid input, from which the network learns to restore the sharp output. We conduct extensive experiments on GoPro dataset and a real event blurry dataset captured by DAVIS240C. Our experimental results on both synthetic and real images demonstrate state-of-the-art performance for <inline-formula> <tex-math notation="LaTeX">$1280\times 720 $ </tex-math></inline-formula> images at 30 fps.
format	article
author	Limeng Zhang Hongguang Zhang Jihua Chen Lei Wang
author_facet	Limeng Zhang Hongguang Zhang Jihua Chen Lei Wang
author_sort	Limeng Zhang
title	Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera
title_short	Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera
title_full	Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera
title_fullStr	Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera
title_full_unstemmed	Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera
title_sort	hybrid deblur net: deep non-uniform deblurring with event camera
publisher	IEEE
publishDate	2020
url	https://doaj.org/article/a23fb01e75674ee2b0894170b13cb055
work_keys_str_mv	AT limengzhang hybriddeblurnetdeepnonuniformdeblurringwitheventcamera AT hongguangzhang hybriddeblurnetdeepnonuniformdeblurringwitheventcamera AT jihuachen hybriddeblurnetdeepnonuniformdeblurringwitheventcamera AT leiwang hybriddeblurnetdeepnonuniformdeblurringwitheventcamera
_version_	1718420684574556160

Hybrid Deblur Net: Deep Non-Uniform Deblurring With Event Camera

Similar Items