Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/134953
Title: DATA-DRIVEN APPROACHES TO UNDERSTANDING VISUAL ATTENTION
Authors: JIANG MING
Keywords: visual saliency,visual attention,eye tracking,machine learning,dataset,autism spectrum disorder
Issue Date: 18-Aug-2016
Source: JIANG MING (2016-08-18). DATA-DRIVEN APPROACHES TO UNDERSTANDING VISUAL ATTENTION. ScholarBank@NUS Repository.
Abstract: The mechanism of visual attention can drastically reduce the amount of information to be processed by more complex perceptual tasks at higher levels, leading to a considerable speed up of visual information processing. In this thesis, we demonstrate that data-driven models of visual attention not only provide interesting insights into how people allocate their attention to different image features of interest, but also help in various scientific and engineering applications. We first attempt to crowd-source the annotation of visual attention on larger-scale image datasets with a novel mouse-tracking paradigm that can be deployed over the Internet. We also present two types of computational models to predict human eye movements as spatial distribution maps and temporal sequences of fixations, respectively. Finally, we quantify the visual attention of people with autism spectrum disorder (ASD) by analyzing their eye movements.
URI: http://scholarbank.nus.edu.sg/handle/10635/134953
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
JiangM.pdf18.19 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

39
checked on Jan 14, 2018

Download(s)

78
checked on Jan 14, 2018

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.