Modelling Objects Using Kernel Principal Component Analysis
Year: 2015
Type: article
Abstract: Object detection is a technologically challenging and practically useful field of computer vision.The success of object detection relies on modelling of an object class. Statistical shape modelling is one of the popular method. Object modelling starts with asset of examples shapes (the training set), and learn from this the pattern of variability of the shape of the class of objects for which the training set can be considered a representative sample. Modelling can considered as the process of modelling the distribution of the training points in shape space. In this paper we present Kernel principal component analysis (KPCA) based active shape models (ASM) for learning the intra –class deformation modes of an object. KPCA is the non-linear dimensionality reduction method. The comparison on performance and space of KPCA and principal component analysis (PCA) are shown Keywords: Object model, KPCA, PCA, ASM. Cite as: Rajkumari Bidyalakshmi Devi, Romesh Laishram, Y.J. Singh, “Modelling Objects Using Kernel Principal Component Analysis” ADBU J.Engg.Tech.,2(1)(2015) 0021102(5pp)
Institution: Assam Don Bosco University
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: diamond
OpenAlex ID: W2302372270