Context-Aware Saliency Detection
Abstract Results Software
Abstract
We propose a new type of saliency - context-aware saliency - which aims at detecting the image regions that represent the scene. This definition differs from previous definitions whose goal is to either identify fixation points or detect the dominant object. In accordance with our saliency definition, we present a detection algorithm which is based on four principles observed in the psychological literature. The benefits of the proposed approach are evaluated in two applications where the context of the dominant objects is just as essential as the objects themselves. In image retargeting we demonstrate that using our saliency prevents distortions in the important regions. In summarization we show that our saliency helps to produce compact, appealing, and informative summaries.
PAMI 2012
CVPR 2010
Results
  

  

Software
The technology referred to in this project, paper and code (below) is protected by a US patent.

This code is based on the following papers (with some small modifications):

  1. Stas Goferman, Lihi Zelnik-Manor, Ayellet Tal, "Context-Aware Saliency Detection", CVPR 2010.
  2. Stas Goferman, Lihi Zelnik-Manor, Ayellet Tal, "Context-Aware Saliency Detection", PAMI 2012. (To Appear)
Please cite this work if you use our software

The software runs on MATLAB version 7.5 (Release R2007b) and up.

The code in this website is for demo purposes only. Individuals or academic institutes are free to use the saliency maps generated using this version as long as they acknowledge its use. Commercial licensing is managed by the Technion Industry Liaison Office. Please contact Hovav Gazit for details.

Lab of CG&M Lab Home Page    Contact Us EE Department Home Page Technion Home Page