A new bottom-up method for saliency detection

Abstract

In this paper, we propose a new method for saliency detection. Unlike other algorithms, our method considers nonsalient regions first, and extracts information in the wavelet transform domain only. Our method need not perform the inverse transform back to the image domain, which can greatly reduce the computational requirements. Experimental results show that our method can achieve excellent results when compared to the state-of-the-art methods.