Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer

Present-day deep learning-based motion deblurring methods utilize the pair of synthetic blur and sharp data to regress any particular framework. This task is designed for directly translating a blurry image input into its restored version as output. The aforementioned approach relies heavily on the...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Jonathan Samuel Lumentut, In Kyu Park
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Motion deblur pseudo-blur augmentation synthesize generative adversarial network human motion Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/b08426708d494ac8ad7d90c8516b91c3
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:b08426708d494ac8ad7d90c8516b91c3
record_format	dspace
spelling	oai:doaj.org-article:b08426708d494ac8ad7d90c8516b91c32021-11-09T00:02:03ZHuman and Scene Motion Deblurring Using Pseudo-Blur Synthesizer2169-353610.1109/ACCESS.2021.3123059https://doaj.org/article/b08426708d494ac8ad7d90c8516b91c32021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9585465/https://doaj.org/toc/2169-3536Present-day deep learning-based motion deblurring methods utilize the pair of synthetic blur and sharp data to regress any particular framework. This task is designed for directly translating a blurry image input into its restored version as output. The aforementioned approach relies heavily on the quality of the synthetic blurry data, which are only available before the training stage. Handling this issue by providing a large amount of data is expensive for common usage. We answer this challenge by providing an on- the-fly blurry data augmenter that can be run during training and test stages. To fully utilize it, we incorporate an unorthodox scheme of deblurring framework that employs the sequence of blur-deblur-reblur-deblur steps. The reblur step is assisted by a reblurring module (synthesizer) that provides the reblurred version (pseudo-blur) of its sharp or deblurred counterpart. The proposed module is also equipped with hand-crafted prior extracted using the state-of-the-art human body statistical model. This prior is employed to map human and non-human regions during adversarial learning to fully perceive the characteristics of human-articulated and scene motion blurs. By engaging this approach, our deblurring module becomes adaptive and achieves superior outcomes compared to recent state-of-the-art deblurring algorithms.Jonathan Samuel LumentutIn Kyu ParkIEEEarticleMotion deblurpseudo-bluraugmentationsynthesizegenerative adversarial networkhuman motionElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 146366-146377 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Motion deblur pseudo-blur augmentation synthesize generative adversarial network human motion Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Motion deblur pseudo-blur augmentation synthesize generative adversarial network human motion Electrical engineering. Electronics. Nuclear engineering TK1-9971 Jonathan Samuel Lumentut In Kyu Park Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer
description	Present-day deep learning-based motion deblurring methods utilize the pair of synthetic blur and sharp data to regress any particular framework. This task is designed for directly translating a blurry image input into its restored version as output. The aforementioned approach relies heavily on the quality of the synthetic blurry data, which are only available before the training stage. Handling this issue by providing a large amount of data is expensive for common usage. We answer this challenge by providing an on- the-fly blurry data augmenter that can be run during training and test stages. To fully utilize it, we incorporate an unorthodox scheme of deblurring framework that employs the sequence of blur-deblur-reblur-deblur steps. The reblur step is assisted by a reblurring module (synthesizer) that provides the reblurred version (pseudo-blur) of its sharp or deblurred counterpart. The proposed module is also equipped with hand-crafted prior extracted using the state-of-the-art human body statistical model. This prior is employed to map human and non-human regions during adversarial learning to fully perceive the characteristics of human-articulated and scene motion blurs. By engaging this approach, our deblurring module becomes adaptive and achieves superior outcomes compared to recent state-of-the-art deblurring algorithms.
format	article
author	Jonathan Samuel Lumentut In Kyu Park
author_facet	Jonathan Samuel Lumentut In Kyu Park
author_sort	Jonathan Samuel Lumentut
title	Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer
title_short	Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer
title_full	Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer
title_fullStr	Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer
title_full_unstemmed	Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer
title_sort	human and scene motion deblurring using pseudo-blur synthesizer
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/b08426708d494ac8ad7d90c8516b91c3
work_keys_str_mv	AT jonathansamuellumentut humanandscenemotiondeblurringusingpseudoblursynthesizer AT inkyupark humanandscenemotiondeblurringusingpseudoblursynthesizer
_version_	1718441413461409792

Human and Scene Motion Deblurring Using Pseudo-Blur Synthesizer

Ejemplares similares