An Efficient Saliency Detection Model Based on Wavelet Generalized Lifting
Document Type
Article
Publication Date
2-1-2019
Abstract
Saliency detection refers to the segmentation of all visually conspicuous objects from various backgrounds. The purpose is to produce an object-mask that overlaps the salient regions annotated by human vision. In this paper, we propose an efficient bottom-up saliency detection model based on wavelet generalized lifting. It requires no kernels with implicit assumptions and prior knowledge. Multiscale wavelet analysis is performed on broadly tuned color feature channels to include a wide range of spatial-frequency information. A nonlinear wavelet filter bank is designed to emphasize the wavelet coefficients, and then a saliency map is obtained through linear combination of the enhanced wavelet coefficients. This full-resolution saliency map uniformly highlights multiple salient objects of different sizes and shapes. An object-mask is constructed by the adaptive thresholding scheme on the saliency maps. Experimental results show that the proposed model outperforms the existing state-of-the-art competitors on two benchmark datasets.
Identifier
85052965788 (Scopus)
Publication Title
International Journal of Pattern Recognition and Artificial Intelligence
External Full Text Location
https://doi.org/10.1142/S0218001419540065
ISSN
02180014
Issue
2
Volume
33
Fund Ref
National Science Foundation
Recommended Citation
Zhong, Xin and Shih, Frank Y., "An Efficient Saliency Detection Model Based on Wavelet Generalized Lifting" (2019). Faculty Publications. 7806.
https://digitalcommons.njit.edu/fac_pubs/7806
