Graph-Based Visual Saliency
2006-12-04Advances in Neural Information Processing Systems 19 2006Unverified0· sign in to hype
Jonathan Harel, Christof Koch, Pietro Perona
Unverified — Be the first to reproduce this paper.
ReproduceAbstract
A new bottom-up visual saliency model, Graph-Based Visual Saliency (GBVS), is proposed. It consists of two steps: rst forming activation maps on certain feature channels, and then normalizing them in a way which highlights conspicuity and admits combination with other maps. The model is simple, and biologically plausible insofar as it is naturally parallelized. This model powerfully predicts human xations on 749 variations of 108 natural images, achieving 98% of the ROC area of a human-based control, whereas the classical algorithms of Itti & Koch ([2], [3], [4]) achieve only 84%.