Title of Paper:Saliency detection via joint modeling global shape and local consistency

Hits:

Document Code:j.neucom.2016.10.007

Date of Publication:2017-01-26

Journal:Neurocomputing

Included Journals:SCI

Affiliation of Author(s):School of Information and Communication Engineering

Place of Publication:Netherlands

Discipline:Engineering

First-Level Discipline:Information and Communication Engineering

Volume:222

Page Number:81-90

ISSN No.:0925-2312

Key Words:Saliency detection; Joint modeling; Object shape; Local consistency

Abstract:Saliency detection is the task of locating informative regions in an image, which is a challenging task incomputer vision. In contrast to the existing saliency detection models that focus on either local or global image
property, an effective salient object detection method is introduced based on joint modeling global shape and local consistency. To this end, Restricted Boltzmann Machine (RBM) is utilized to model salient object shape as
global image property and Conditional Random Field (CRF), on the other hand, is adopted to achieve its local consistency. In order to obtain the final salienc