Focus-and-Context Skeleton-based Image Simplification using Saliency Maps

J. Wang

doi:10.5220/0010193400450055

Focus-and-Context Skeleton-based Image Simplification using Saliency Maps

J. Wang

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

Abstract

Medial descriptors offer a promising way for representing, simplifying, manipulating, and compressing images. However, to date, these have been applied in a global manner that is oblivious to salient features. In this paper, we adapt medial descriptors to use the information provided by saliency maps to selectively simplify and encode an image while preserving its salient regions. This allows us to improve the trade-off between compression ratio and image quality as compared to the standard dense-skeleton method while keeping perceptually salient features, in a focus-and-context manner. We show how our method can be combined with JPEG to increase overall compression rates at the cost of a slightly lower image quality. We demonstrate our method on a benchmark composed of a broad set of images.

Original language	English
Title of host publication	Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,
Editors	Giovanni Maria Farinella, Petia Radeva, Jose Braz, Kadi Bouatouch
Publisher	INSTICC
Pages	45-55
Number of pages	11
Volume	4
ISBN (Electronic)	9789897584886
ISBN (Print)	978-989-758-488-6
DOIs	https://doi.org/10.5220/0010193400450055
Publication status	Published - 2021
Event	16th International Conference on Computer Graphics Theory and Applications - Duration: 8 Feb 2021 → 10 Feb 2021

Conference

Conference	16th International Conference on Computer Graphics Theory and Applications
Abbreviated title	VISIGRAPP 2021
Period	8/02/21 → 10/02/21

Bibliographical note

Funding Information:
The first author acknowledges the China Scholarship Council (Grant number: 201806320354) for financial support.

Publisher Copyright:
Copyright © 2021 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.

Keywords

Dense Skeleton
Image Simplification
Medial Axis
Saliency Map

Access to Document

10.5220/0010193400450055Licence: CC BY-NC-ND

78923158_6196932_ExampleFinal published version, 2.96 MBLicence: CC BY-NC-ND

Cite this

Wang, J. (2021). Focus-and-Context Skeleton-based Image Simplification using Saliency Maps. In G. M. Farinella, P. Radeva, J. Braz, & K. Bouatouch (Eds.), Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, (Vol. 4, pp. 45-55). INSTICC. https://doi.org/10.5220/0010193400450055

@inproceedings{b84ff7356aea40f5822fb101d7ec6de7,

title = "Focus-and-Context Skeleton-based Image Simplification using Saliency Maps",

abstract = "Medial descriptors offer a promising way for representing, simplifying, manipulating, and compressing images. However, to date, these have been applied in a global manner that is oblivious to salient features. In this paper, we adapt medial descriptors to use the information provided by saliency maps to selectively simplify and encode an image while preserving its salient regions. This allows us to improve the trade-off between compression ratio and image quality as compared to the standard dense-skeleton method while keeping perceptually salient features, in a focus-and-context manner. We show how our method can be combined with JPEG to increase overall compression rates at the cost of a slightly lower image quality. We demonstrate our method on a benchmark composed of a broad set of images.",

keywords = "Dense Skeleton, Image Simplification, Medial Axis, Saliency Map",

author = "J. Wang",

note = "Funding Information: The first author acknowledges the China Scholarship Council (Grant number: 201806320354) for financial support. Publisher Copyright: Copyright {\textcopyright} 2021 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.; 16th International Conference on Computer Graphics Theory and Applications, VISIGRAPP 2021 ; Conference date: 08-02-2021 Through 10-02-2021",

year = "2021",

doi = "10.5220/0010193400450055",

language = "English",

isbn = "978-989-758-488-6",

volume = "4",

pages = "45--55",

editor = "Farinella, \{Giovanni Maria\} and Petia Radeva and Jose Braz and Kadi Bouatouch",

booktitle = "Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,",

publisher = "INSTICC",

}

Wang, J 2021, Focus-and-Context Skeleton-based Image Simplification using Saliency Maps. in GM Farinella, P Radeva, J Braz & K Bouatouch (eds), Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,. vol. 4, INSTICC, pp. 45-55, 16th International Conference on Computer Graphics Theory and Applications, 8/02/21. https://doi.org/10.5220/0010193400450055

Focus-and-Context Skeleton-based Image Simplification using Saliency Maps. / Wang, J.
Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,. ed. / Giovanni Maria Farinella; Petia Radeva; Jose Braz; Kadi Bouatouch. Vol. 4 INSTICC, 2021. p. 45-55.

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

TY - GEN

T1 - Focus-and-Context Skeleton-based Image Simplification using Saliency Maps

AU - Wang, J.

N1 - Funding Information: The first author acknowledges the China Scholarship Council (Grant number: 201806320354) for financial support. Publisher Copyright: Copyright © 2021 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved.

PY - 2021

Y1 - 2021

N2 - Medial descriptors offer a promising way for representing, simplifying, manipulating, and compressing images. However, to date, these have been applied in a global manner that is oblivious to salient features. In this paper, we adapt medial descriptors to use the information provided by saliency maps to selectively simplify and encode an image while preserving its salient regions. This allows us to improve the trade-off between compression ratio and image quality as compared to the standard dense-skeleton method while keeping perceptually salient features, in a focus-and-context manner. We show how our method can be combined with JPEG to increase overall compression rates at the cost of a slightly lower image quality. We demonstrate our method on a benchmark composed of a broad set of images.

AB - Medial descriptors offer a promising way for representing, simplifying, manipulating, and compressing images. However, to date, these have been applied in a global manner that is oblivious to salient features. In this paper, we adapt medial descriptors to use the information provided by saliency maps to selectively simplify and encode an image while preserving its salient regions. This allows us to improve the trade-off between compression ratio and image quality as compared to the standard dense-skeleton method while keeping perceptually salient features, in a focus-and-context manner. We show how our method can be combined with JPEG to increase overall compression rates at the cost of a slightly lower image quality. We demonstrate our method on a benchmark composed of a broad set of images.

KW - Dense Skeleton

KW - Image Simplification

KW - Medial Axis

KW - Saliency Map

UR - https://www.scopus.com/pages/publications/85103034529

U2 - 10.5220/0010193400450055

DO - 10.5220/0010193400450055

M3 - Conference contribution

SN - 978-989-758-488-6

VL - 4

SP - 45

EP - 55

BT - Proceedings of the 16th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,

A2 - Farinella, Giovanni Maria

A2 - Radeva, Petia

A2 - Braz, Jose

A2 - Bouatouch, Kadi

PB - INSTICC

T2 - 16th International Conference on Computer Graphics Theory and Applications

Y2 - 8 February 2021 through 10 February 2021

ER -

Focus-and-Context Skeleton-based Image Simplification using Saliency Maps

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this