Saliency Detection with Spaces of Background-based Distribution

TL;DR

This paper introduces a novel saliency detection method that models background distributions using eigendecomposition and Mahalanobis distance, resulting in improved saliency maps through Bayesian and geodesic refinement.

Contribution

It proposes a new background modeling approach with four eigendecomposition-based spaces and a refined saliency detection pipeline combining Bayesian and geodesic methods.

Findings

Outperforms state-of-the-art methods on benchmark datasets.

Effective background modeling improves saliency detection accuracy.

Combines multiple distance measures for robust saliency estimation.

Abstract

In this letter, an effective image saliency detection method is proposed by constructing some novel spaces to model the background and redefine the distance of the salient patches away from the background. Concretely, given the backgroundness prior, eigendecomposition is utilized to create four spaces of background-based distribution (SBD) to model the background, in which a more appropriate metric (Mahalanobis distance) is quoted to delicately measure the saliency of every image patch away from the background. After that, a coarse saliency map is obtained by integrating the four adjusted Mahalanobis distance maps, each of which is formed by the distances between all the patches and background in the corresponding SBD. To be more discriminative, the coarse saliency map is further enhanced into the posterior probability map within Bayesian perspective. Finally, the final saliency map is generated by properly refining the posterior probability map with geodesic distance. Experimental results on two usual datasets show that the proposed method is effective compared with the state-of-the-art algorithms.