Hershey, Pennsylvania
New York, New YorkBeijing, China
Machine Learning Applications in Computer Vision

Machine Learning Applications in Computer Vision

Mehrtash Harandi (NICTA, Australia & The University of Queensland, Australia), Javid Taheri (The University of Sydney, Australia), and Brian C. Lovell (NICTA, Australia & The University of Queensland, Australia)
Copyright: © 2012 |Pages: 34
DOI: 10.4018/978-1-4666-1833-6.ch007
OnDemand:
(Individual Chapters)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Recognizing objects based on their appearance (visual recognition) is one of the most significant abilities of many living creatures. In this study, recent advances in the area of automated object recognition are reviewed; the authors specifically look into several learning frameworks to discuss how they can be utilized in solving object recognition paradigms. This includes reinforcement learning, a biologically-inspired machine learning technique to solve sequential decision problems and transductive learning, and a framework where the learner observes query data and potentially exploits its structure for classification. The authors also discuss local and global appearance models for object recognition, as well as how similarities between objects can be learnt and evaluated.
Chapter Preview
Top

1- Object Description And Representation In Computer Vision

Objects can be described by different cues. For example, objects can be described by geometrical primitives like boxes, spheres, cones, and cylinders. Describing and representing objects based on their appearance is a widely studied approach in the literature (Bartlett, Movellan, & Sejnowski, 2002; Belhumeur, Hespanha, & Kriegman, 1997; Dorko & Schmid, 2005; Lowe, 2004; Mikolajczyk, Leibe, & Schiele, 2005; Serre, Wolf, Bileschi, Riesenhuber, & Poggio, 2007; Shechtman & Irani, 2007; Turk & Pentland, 1991; H. Zhang, Gao, Chen, & Zhao, 2006). The general idea of appearance-based object recognition is to extract useful and robust information from only the appearance of the object-of-interest that is usually captured by different two-dimensional views. Appearance-based methods can be sub-divided into two main classes: local and global approaches.

Complete Chapter List

Search this Book:
Reset