Non-uniform deblur using gyro sensor and long/short exposure image pair

Seung Ji Seo; Ho Hyoung Ryu; Dongyun Choi; Byung Cheol Song

doi:10.1007/978-3-319-24075-6_3

Non-uniform deblur using gyro sensor and long/short exposure image pair

Seung Ji Seo, Ho Hyoung Ryu, Dongyun Choi, Byung Cheol Song

Inha University

Research output: Contribution to journal › Conference article › peer-review

Abstract

This paper proposes a deblur algorithm using IMU sensor and long/ short exposure-time image pair. First, we derive an initial blur kernel from gyro data of IMU sensor. Second, we refine the blur kernel by applying Lucas-Kanade algorithm to long/short exposure-time image pair. Using residual deconvolution based on the non-uniform blur kernel, we synthesize a final image. Experimental results show that the proposed algorithm is superior to the state-of-the-art methods in terms of subjective/objective visual quality.

Original language	English
Pages (from-to)	23-31
Number of pages	9
Journal	Lecture Notes in Computer Science
Volume	9314
DOIs	https://doi.org/10.1007/978-3-319-24075-6_3
State	Published - 2015
Event	16th Pacific-Rim Conference on Multimedia, PCM 2015 - Gwangju, Korea, Republic of Duration: 16 Sep 2015 → 18 Sep 2015

Bibliographical note

Publisher Copyright:
© Springer International Publishing Switzerland 2015.

Keywords

Deblurring
Gyro
IMU
Non-uniform blur

Access to Document

10.1007/978-3-319-24075-6_3

Cite this

@article{5b46da87721746cb930af8051e37e14a,

title = "Non-uniform deblur using gyro sensor and long/short exposure image pair",

abstract = "This paper proposes a deblur algorithm using IMU sensor and long/ short exposure-time image pair. First, we derive an initial blur kernel from gyro data of IMU sensor. Second, we refine the blur kernel by applying Lucas-Kanade algorithm to long/short exposure-time image pair. Using residual deconvolution based on the non-uniform blur kernel, we synthesize a final image. Experimental results show that the proposed algorithm is superior to the state-of-the-art methods in terms of subjective/objective visual quality.",

keywords = "Deblurring, Gyro, IMU, Non-uniform blur",

author = "Seo, \{Seung Ji\} and Ryu, \{Ho Hyoung\} and Dongyun Choi and Song, \{Byung Cheol\}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2015.; 16th Pacific-Rim Conference on Multimedia, PCM 2015 ; Conference date: 16-09-2015 Through 18-09-2015",

year = "2015",

doi = "10.1007/978-3-319-24075-6\_3",

language = "English",

volume = "9314",

pages = "23--31",

journal = "Lecture Notes in Computer Science",

issn = "0302-9743",

publisher = "Springer Science and Business Media Deutschland GmbH",

}

TY - JOUR

T1 - Non-uniform deblur using gyro sensor and long/short exposure image pair

AU - Seo, Seung Ji

AU - Ryu, Ho Hyoung

AU - Choi, Dongyun

AU - Song, Byung Cheol

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2015.

PY - 2015

Y1 - 2015

N2 - This paper proposes a deblur algorithm using IMU sensor and long/ short exposure-time image pair. First, we derive an initial blur kernel from gyro data of IMU sensor. Second, we refine the blur kernel by applying Lucas-Kanade algorithm to long/short exposure-time image pair. Using residual deconvolution based on the non-uniform blur kernel, we synthesize a final image. Experimental results show that the proposed algorithm is superior to the state-of-the-art methods in terms of subjective/objective visual quality.

AB - This paper proposes a deblur algorithm using IMU sensor and long/ short exposure-time image pair. First, we derive an initial blur kernel from gyro data of IMU sensor. Second, we refine the blur kernel by applying Lucas-Kanade algorithm to long/short exposure-time image pair. Using residual deconvolution based on the non-uniform blur kernel, we synthesize a final image. Experimental results show that the proposed algorithm is superior to the state-of-the-art methods in terms of subjective/objective visual quality.

KW - Deblurring

KW - Gyro

KW - IMU

KW - Non-uniform blur

UR - https://www.scopus.com/pages/publications/84984640084

U2 - 10.1007/978-3-319-24075-6_3

DO - 10.1007/978-3-319-24075-6_3

M3 - Conference article

AN - SCOPUS:84984640084

SN - 0302-9743

VL - 9314

SP - 23

EP - 31

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

T2 - 16th Pacific-Rim Conference on Multimedia, PCM 2015

Y2 - 16 September 2015 through 18 September 2015

ER -

Non-uniform deblur using gyro sensor and long/short exposure image pair

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this