Image stitching using multiple homographies estimated by segmented regions for different parallaxes

Daeho Lee; Jaeyoung Yoon; Sunggyel Lim

doi:10.1109/ICVISP.2017.19

Image stitching using multiple homographies estimated by segmented regions for different parallaxes

Daeho Lee, Jaeyoung Yoon, Sunggyel Lim

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

5 Citations (Scopus)

Abstract

We propose a novel image stitching method using multiple homographies. The method can stitch images having different parallaxes, such as an image that contains distant buildings and close trees. Such images might not be stitched with fine quality by single global homography. Therefore, we select a background homography by analyzing the inliers of the homographies estimated by RANSAC (random sample consensus) for leave-one-out segmented regions, and the left segmented region of the background homography is designated as an object region. Then, the object homography is estimated by the object region. Using these multiple homographies, image stitching and multi-band blending are performed. With this method, images having different parallaxes are stitched with higher visual quality than with other methods using single homography and multiple homographies.

Original language	English
Title of host publication	Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	71-75
Number of pages	5
ISBN (Electronic)	9781538606124
DOIs	https://doi.org/10.1109/ICVISP.2017.19
Publication status	Published - 28 Nov 2017
Event	2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017 - Osaka, Japan Duration: 22 Sept 2017 → 24 Sept 2017

Publication series

Name	Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017
Volume	2017-November

Conference

Conference	2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017
Country/Territory	Japan
City	Osaka
Period	22/09/17 → 24/09/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Keywords

Different parallaxes
Image stitching
Multiple homographies

Access to Document

10.1109/ICVISP.2017.19

Cite this

Lee, D., Yoon, J., & Lim, S. (2017). Image stitching using multiple homographies estimated by segmented regions for different parallaxes. In Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017 (pp. 71-75). (Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017; Vol. 2017-November). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICVISP.2017.19

Lee, Daeho ; Yoon, Jaeyoung ; Lim, Sunggyel. / Image stitching using multiple homographies estimated by segmented regions for different parallaxes. Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 71-75 (Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017).

@inproceedings{76919accf64245c98b7c333d3e227163,

title = "Image stitching using multiple homographies estimated by segmented regions for different parallaxes",

abstract = "We propose a novel image stitching method using multiple homographies. The method can stitch images having different parallaxes, such as an image that contains distant buildings and close trees. Such images might not be stitched with fine quality by single global homography. Therefore, we select a background homography by analyzing the inliers of the homographies estimated by RANSAC (random sample consensus) for leave-one-out segmented regions, and the left segmented region of the background homography is designated as an object region. Then, the object homography is estimated by the object region. Using these multiple homographies, image stitching and multi-band blending are performed. With this method, images having different parallaxes are stitched with higher visual quality than with other methods using single homography and multiple homographies.",

keywords = "Different parallaxes, Image stitching, Multiple homographies",

author = "Daeho Lee and Jaeyoung Yoon and Sunggyel Lim",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.; 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017 ; Conference date: 22-09-2017 Through 24-09-2017",

year = "2017",

month = nov,

day = "28",

doi = "10.1109/ICVISP.2017.19",

language = "English",

series = "Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "71--75",

booktitle = "Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017",

address = "United States",

}

Lee, D, Yoon, J & Lim, S 2017, Image stitching using multiple homographies estimated by segmented regions for different parallaxes. in Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017. Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017, vol. 2017-November, Institute of Electrical and Electronics Engineers Inc., pp. 71-75, 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017, Osaka, Japan, 22/09/17. https://doi.org/10.1109/ICVISP.2017.19

Image stitching using multiple homographies estimated by segmented regions for different parallaxes. / Lee, Daeho; Yoon, Jaeyoung; Lim, Sunggyel.
Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017. Institute of Electrical and Electronics Engineers Inc., 2017. p. 71-75 (Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017; Vol. 2017-November).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Image stitching using multiple homographies estimated by segmented regions for different parallaxes

AU - Lee, Daeho

AU - Yoon, Jaeyoung

AU - Lim, Sunggyel

PY - 2017/11/28

Y1 - 2017/11/28

N2 - We propose a novel image stitching method using multiple homographies. The method can stitch images having different parallaxes, such as an image that contains distant buildings and close trees. Such images might not be stitched with fine quality by single global homography. Therefore, we select a background homography by analyzing the inliers of the homographies estimated by RANSAC (random sample consensus) for leave-one-out segmented regions, and the left segmented region of the background homography is designated as an object region. Then, the object homography is estimated by the object region. Using these multiple homographies, image stitching and multi-band blending are performed. With this method, images having different parallaxes are stitched with higher visual quality than with other methods using single homography and multiple homographies.

AB - We propose a novel image stitching method using multiple homographies. The method can stitch images having different parallaxes, such as an image that contains distant buildings and close trees. Such images might not be stitched with fine quality by single global homography. Therefore, we select a background homography by analyzing the inliers of the homographies estimated by RANSAC (random sample consensus) for leave-one-out segmented regions, and the left segmented region of the background homography is designated as an object region. Then, the object homography is estimated by the object region. Using these multiple homographies, image stitching and multi-band blending are performed. With this method, images having different parallaxes are stitched with higher visual quality than with other methods using single homography and multiple homographies.

KW - Different parallaxes

KW - Image stitching

KW - Multiple homographies

UR - http://www.scopus.com/inward/record.url?scp=85048430873&partnerID=8YFLogxK

U2 - 10.1109/ICVISP.2017.19

DO - 10.1109/ICVISP.2017.19

M3 - Conference contribution

AN - SCOPUS:85048430873

T3 - Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017

SP - 71

EP - 75

BT - Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017

Y2 - 22 September 2017 through 24 September 2017

ER -

Lee D, Yoon J, Lim S. Image stitching using multiple homographies estimated by segmented regions for different parallaxes. In Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017. Institute of Electrical and Electronics Engineers Inc. 2017. p. 71-75. (Proceedings - 2017 International Conference on Vision, Image and Signal Processing, ICVISP 2017). doi: 10.1109/ICVISP.2017.19

Image stitching using multiple homographies estimated by segmented regions for different parallaxes

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this