Skip to main content

Research Repository

Advanced Search

A low complexity visual saliency model based on in-focus regions and centre sensitivity.

Chilukamari, Jayachandra; Kannangara, Sampath; Maxwell, Grant

Authors

Jayachandra Chilukamari

Sampath Kannangara

Grant Maxwell



Abstract

A novel low-complexity visual saliency detection algorithm for detecting salient regions in images is proposed. The algorithm derives salient regions based on in-focus regions and image centre sensitivity. The performance of the algorithm in predicting human eye fixations is validated against ten state-of-the-art algorithms using a public image dataset. The results demonstrate that the proposed algorithm achieves higher prediction accuracy in saliency detection at significantly lower computational complexity compared to other algorithms.

Citation

CHILUKAMARI, J., KANNANGARA, S. and MAXWELL, G. 2014. A low complexity visual saliency model based on in-focus regions and centre sensitivity. In Proceedings of the 4th IEEE international conference on consumer electronics (ICCE-Berlin 2014), 7-10 September 2014, Berlin, Germany. New York: IEEE [online], article number 7034250, pages 411-414. Available from: https://doi.org/10.1109/ICCE-Berlin.2014.7034250

Conference Name 4th IEEE international conference on consumer electronics (ICCE-Berlin 2014)
Conference Location Berlin, Germany
Start Date Sep 7, 2014
End Date Sep 10, 2014
Acceptance Date May 22, 2014
Online Publication Date Sep 7, 2014
Publication Date Feb 9, 2015
Deposit Date Aug 29, 2017
Publicly Available Date Aug 29, 2017
Publisher Institute of Electrical and Electronics Engineers
Article Number 7034250
Pages 411-414
ISBN 9781479961665
DOI https://doi.org/10.1109/ICCE-Berlin.2014.7034250
Keywords Visual saliency; Visual attention models; Saliency detection
Public URL http://hdl.handle.net/10059/2479
Related Public URLs http://hdl.handle.net/10059/2443 ; http://hdl.handle.net/10059/2477 ; http://hdl.handle.net/10059/2478 ; http://hdl.handle.net/10059/2481

Files





You might also like



Downloadable Citations