A Model of Space and Object-based Attention for Visual Saliency

In this paper, we proposed a model for visual saliency, which is space and object based and can be used for detecting single and multiple salient objects in static natural scenes. The model uses a bottom-up saliency map based on color, orientation and intensity, integrated with edge areas based on e...

Full description

Saved in:
Bibliographic Details
Main Authors: Jingjing Zhong, Siwei Luo
Format: Conference Proceeding
Language:English
Subjects:
Color
Detectors
Humans
Information processing
Information technology
Layout
Object detection
Space technology
Sun
Visual system
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this paper, we proposed a model for visual saliency, which is space and object based and can be used for detecting single and multiple salient objects in static natural scenes. The model uses a bottom-up saliency map based on color, orientation and intensity, integrated with edge areas based on edge information. Edge areas is proposed here first as we know, and most likely to be object regions. It gets from the convex hulls of edge groups, and fits the objects well in many cases. The results of the proposed model on natural scene are reported
DOI:10.1109/ICICIC.2006.202