Please use this identifier to cite or link to this item: http://scholarbank.nus.edu.sg/handle/10635/34470
Title: Local Features to a Global View: Recognition of Occluded Objects by Spectral Matching Using Pairwise Feature Relationships
Authors: WU JIAYUN
Keywords: occlusion, object recognition, spectral matching, feature relationships
Issue Date: 29-Feb-2012
Source: WU JIAYUN (2012-02-29). Local Features to a Global View: Recognition of Occluded Objects by Spectral Matching Using Pairwise Feature Relationships. ScholarBank@NUS Repository.
Abstract: In this research, our algorithms to solve the recognition of occluded object problem are formulated by a local to global strategy, namely making a recognition decision (global) based on local information collected. This local to global nature of occlusion recognition has brought us to spectral matching theory for its ability to determine global structural properties of graphs. However, spectral algorithms respond badly to contaminated data set, such as occluded objects, where ambiguous connections are generated. Therefore, our efforts are focused on how to reduce interactions of features from different objects. Reducing feature interactions for spectral correspondence is the key of our algorithms. We propose feature interaction reduction schemes based on intermediate level vision cues: grouping and segmentation, integrating low-level, intermediate-level into higher level vision tasks. With our feature interaction formulations, inter and intra feature relationships are established to indicate their possibility to come from the same object.
URI: http://scholarbank.nus.edu.sg/handle/10635/34470
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
WuJY.pdf3.06 MBAdobe PDF

OPEN

NoneView/Download

Page view(s)

230
checked on Dec 18, 2017

Download(s)

412
checked on Dec 18, 2017

Google ScholarTM

Check


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