Learning optimal seeds for diffusion-based salient object detection

In diffusion-based saliency detection, an image is parti-tioned into superpixels and mapped to a graph, with super-pixels as nodes and edge strengths proportional to super-pixel similarity. Saliency information is then propagated over the graph using a diffusion process, whose equilibrium state yields the object saliency map. The optimal solution is the product of a propagation matrix and a saliency seed vector that contains a prior saliency assessment. This is obtained from either a bottom-up saliency detector or some heuristics. In this work, we propose a method to learn op-timal seeds for object saliency. Two types of features are computed per superpixel: the bottom-up saliency of the su-perpixel region and a set of mid-level vision features infor-mative of how likely the superpixel is to belong to an object. The combination of features that best discriminates between object and background saliency is then learned, using a large-margin formulation of the discriminant saliency prin-ciple. The propagation of the resulting saliency seeds, using a diffusion process, is finally shown to outperform the state of the art on a number of salient object detection datasets. 1