A parametric spectral model for texture-based salience

Kasim Terzić; Sai Krishna; J. M.H. Du Buf

doi:10.1007/978-3-319-24947-6_27

A parametric spectral model for texture-based salience

Kasim Terzić^*, Sai Krishna, J. M.H. Du Buf

^*Corresponding author for this work

School of Computer Science

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

Abstract

We present a novel saliency mechanism based on texture. Local texture at each pixel is characterised by the 2D spectrum obtained from oriented Gabor filters. We then apply a parametric model and describe the texture at each pixel by a combination of two 1D Gaussian approximations. This results in a simple model which consists of only four parameters. These four parameters are then used as feature channels and standard Difference-of-Gaussian blob detection is applied in order to detect salient areas in the image, similar to the Itti and Koch model. Finally, a diffusion process is used to sharpen the resulting regions. Evaluation on a large saliency dataset shows a significant improvement of our method over the baseline Itti and Koch model.

Original language	English
Title of host publication	Pattern Recognition
Subtitle of host publication	37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings
Editors	Juergen Gall, Peter Gehler, Bastian Leibe
Place of Publication	Cham
Publisher	Springer
Pages	331-342
Number of pages	12
ISBN (Electronic)	9783319249476
ISBN (Print)	9783319249469
DOIs	https://doi.org/10.1007/978-3-319-24947-6_27
Publication status	Published - 2015
Event	37th German Conference on Pattern Recognition, GCPR 2015 - Aachen, Germany Duration: 7 Oct 2015 → 10 Oct 2015 Conference number: 37 http://gcpr2015.rwth-aachen.de/

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9358
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	37th German Conference on Pattern Recognition, GCPR 2015
Abbreviated title	GCPR
Country/Territory	Germany
City	Aachen
Period	7/10/15 → 10/10/15
Internet address	http://gcpr2015.rwth-aachen.de/

Access to Document

10.1007/978-3-319-24947-6_27

gcprfinal-106
© 2015, Springer International Publishing Switzerland. This work has been made available online in accordance with the publisher’s policies. This is the author created accepted version manuscript following peer review and as such may differ slightly from the final published version. The final published version of this work is available at https://doi.org/10.1007/978-3-319-24947-6_27
Accepted author manuscript, 3.22 MB

Cite this

Terzić, K., Krishna, S., & Du Buf, J. M. H. (2015). A parametric spectral model for texture-based salience. In J. Gall, P. Gehler, & B. Leibe (Eds.), Pattern Recognition : 37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings (pp. 331-342). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9358). Springer. https://doi.org/10.1007/978-3-319-24947-6_27

Terzić, Kasim ; Krishna, Sai ; Du Buf, J. M.H. / A parametric spectral model for texture-based salience. Pattern Recognition : 37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings. editor / Juergen Gall ; Peter Gehler ; Bastian Leibe. Cham : Springer, 2015. pp. 331-342 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{d19955defbce4b83bbf8d88bde7d9c65,

title = "A parametric spectral model for texture-based salience",

abstract = "We present a novel saliency mechanism based on texture. Local texture at each pixel is characterised by the 2D spectrum obtained from oriented Gabor filters. We then apply a parametric model and describe the texture at each pixel by a combination of two 1D Gaussian approximations. This results in a simple model which consists of only four parameters. These four parameters are then used as feature channels and standard Difference-of-Gaussian blob detection is applied in order to detect salient areas in the image, similar to the Itti and Koch model. Finally, a diffusion process is used to sharpen the resulting regions. Evaluation on a large saliency dataset shows a significant improvement of our method over the baseline Itti and Koch model.",

author = "Kasim Terzi{\'c} and Sai Krishna and \{Du Buf\}, \{J. M.H.\}",

year = "2015",

doi = "10.1007/978-3-319-24947-6\_27",

language = "English",

isbn = "9783319249469",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer",

pages = "331--342",

editor = "Juergen Gall and Peter Gehler and Bastian Leibe",

booktitle = "Pattern Recognition",

address = "Netherlands",

note = "37th German Conference on Pattern Recognition, GCPR 2015, GCPR ; Conference date: 07-10-2015 Through 10-10-2015",

url = "http://gcpr2015.rwth-aachen.de/",

}

Terzić, K, Krishna, S & Du Buf, JMH 2015, A parametric spectral model for texture-based salience. in J Gall, P Gehler & B Leibe (eds), Pattern Recognition : 37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9358, Springer, Cham, pp. 331-342, 37th German Conference on Pattern Recognition, GCPR 2015, Aachen, North Rhine-Westphalia, Germany, 7/10/15. https://doi.org/10.1007/978-3-319-24947-6_27

A parametric spectral model for texture-based salience. / Terzić, Kasim; Krishna, Sai; Du Buf, J. M.H.
Pattern Recognition : 37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings. ed. / Juergen Gall; Peter Gehler; Bastian Leibe. Cham: Springer, 2015. p. 331-342 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9358).

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

TY - GEN

T1 - A parametric spectral model for texture-based salience

AU - Terzić, Kasim

AU - Krishna, Sai

AU - Du Buf, J. M.H.

N1 - Conference code: 37

PY - 2015

Y1 - 2015

N2 - We present a novel saliency mechanism based on texture. Local texture at each pixel is characterised by the 2D spectrum obtained from oriented Gabor filters. We then apply a parametric model and describe the texture at each pixel by a combination of two 1D Gaussian approximations. This results in a simple model which consists of only four parameters. These four parameters are then used as feature channels and standard Difference-of-Gaussian blob detection is applied in order to detect salient areas in the image, similar to the Itti and Koch model. Finally, a diffusion process is used to sharpen the resulting regions. Evaluation on a large saliency dataset shows a significant improvement of our method over the baseline Itti and Koch model.

AB - We present a novel saliency mechanism based on texture. Local texture at each pixel is characterised by the 2D spectrum obtained from oriented Gabor filters. We then apply a parametric model and describe the texture at each pixel by a combination of two 1D Gaussian approximations. This results in a simple model which consists of only four parameters. These four parameters are then used as feature channels and standard Difference-of-Gaussian blob detection is applied in order to detect salient areas in the image, similar to the Itti and Koch model. Finally, a diffusion process is used to sharpen the resulting regions. Evaluation on a large saliency dataset shows a significant improvement of our method over the baseline Itti and Koch model.

UR - https://www.scopus.com/pages/publications/84952359587

U2 - 10.1007/978-3-319-24947-6_27

DO - 10.1007/978-3-319-24947-6_27

M3 - Conference contribution

AN - SCOPUS:84952359587

SN - 9783319249469

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 331

EP - 342

BT - Pattern Recognition

A2 - Gall, Juergen

A2 - Gehler, Peter

A2 - Leibe, Bastian

PB - Springer

CY - Cham

T2 - 37th German Conference on Pattern Recognition, GCPR 2015

Y2 - 7 October 2015 through 10 October 2015

ER -

Terzić K, Krishna S, Du Buf JMH. A parametric spectral model for texture-based salience. In Gall J, Gehler P, Leibe B, editors, Pattern Recognition : 37th German Conference, GCPR 2015, Aachen, Germany, October 7-10, 2015, Proceedings. Cham: Springer. 2015. p. 331-342. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). Epub 2015 Nov 3. doi: 10.1007/978-3-319-24947-6_27

A parametric spectral model for texture-based salience

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this