Title :
A new approach to object recognition for multi viewed objects
Author :
Dalai, Sovan ; Rana, Deepak
Author_Institution :
Dept. of ECE, Centurion Univ. of Technol. & Manage., Bhubaneswar, India
Abstract :
In this paper, we proposed a new method for object recognition for multi viewed object based on shape space. Here we extract the similar features of all the objects of different classes and identified in a single point in a high dimensional manifold which is known as shape space. Object recognition is achieved by measuring the Euclidian distance between an observed object and a model in the shape space. The problem of image retrieval is much more challenging than that of text retrieval since significant features for representing and searching an arbitrary image vary with the image content. So, we focus on landmarks of processing images which are easily recognisable by human users.
Keywords :
feature extraction; image retrieval; object recognition; Euclidian distance; arbitrary image; feature extraction; high dimensional manifold; image content; image retrieval; landmark selection; multi viewed objects; object recognition; observed object; shape space based approach; text retrieval; Computer vision; Databases; Image edge detection; Object recognition; Polynomials; Shape; Euclidean distance; landmark selection; shape space;
Conference_Titel :
Communication, Information & Computing Technology (ICCICT), 2012 International Conference on
Conference_Location :
Mumbai
Print_ISBN :
978-1-4577-2077-2
DOI :
10.1109/ICCICT.2012.6398165
