Abstract
The goal of saliency detection is to locate the regions which are most likely to capture human׳s attention without prior knowledge of their contents. Visual saliency detection has been widely used in image processing, but it is still a challenging problem in computer vision. In this paper, we propose a salient region detection algorithm by integrating global features namely uniqueness and spatial distribution. Two measures of contrast are computed in pixel and superpixel level respectively. In order to suppress background noise, Low-level features are refined by High-level priors which are computed with the Gaussian model based on salient region. We formulate salient region detection as a binary labeling problem that separates salient region from the background. A Conditional Random Field is learned to effectively combine these refined features for salient region detection. Experimental results on the large benchmark database demonstrate the proposed method performs well when against fifteen state-of-the-art methods in terms of precision and recall.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call