Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published January 2012 | public
Journal Article

Image Signature: Highlighting Sparse Salient Regions

Abstract

We introduce a simple image descriptor referred to as the image signature. We show, within the theoretical framework of sparse signal mixing, that this quantity spatially approximates the foreground of an image. We experimentally investigate whether this approximate foreground overlaps with visually conspicuous image locations by developing a saliency algorithm based on the image signature. This saliency algorithm predicts human fixation points best among competitors on the Bruce and Tsotsos [1] benchmark data set and does so in much shorter running time. In a related experiment, we demonstrate with a change blindness data set that the distance between images induced by the image signature is closer to human perceptual distance than can be achieved using other saliency algorithms, pixel-wise, or GIST [2] descriptor methods.

Additional Information

© 2012 IEEE. Manuscript received 7 Jan. 2011; revised 21 Apr. 2011; accepted 1 June 2011; published online 19 July 2011. The first author would like to thank Anders Hansen, Xiaodong Li, and Emmanuel Candes for their insightful discussions. We gratefully acknowledge Claudia Wilimzig, Amy Chung-Yu Chou, and Tom Laudes, who generated the images and collected data for the change blindness experiment. The research was supported by the NeoVision program at the US Defense Advanced Research Projects Agency (DARPA), by the US Office of Naval Research (via an award made through Johns Hopkins University), by the G. Harold & Leila Y. Mathers Charitable Foundation, and by the WCU (World Class University) program funded by the Ministry of Education, Science and Technology through the National Research Foundation of Korea (R31-10008).

Additional details

Created:
September 14, 2023
Modified:
October 23, 2023