AUTHORS: Worawat Choensawat, Woong Choi and Kozaburo Hachimura

ABSTRACT: A similarity retrieval of motion capture data has received substantial attention in recent years. In this paper, we focus on feature extraction and quick filtering methods in the similarity retrieval system. A representation of motion capture data is joint angles, which can distinguish different human body poses. We propose a new technique for dimensionality reduction based the average and variance of joint angles. Our dimensionality reduction is simple to

Read more

AUTHORS: Jakarin Smitaveja, Kingkarn Sookhanaphibarn and Chidchanok Lursinsap

ABSTRACT: Face recognition technology has been an increasingly important module in security systems. A challenging problem is how to extract features tolerant to the appearance variables such as changes in shape, illumination, and occlusion. Extracted metrical features of facial caricatures that are combined with their facial photographs in the training set are examined. The facial caricature is a personal representative amplifying perceptually significant information of individuals. Unlike Eigenfaces, Fisherfaces, and Laplacianfaces, the twenty-nine metrical features that used in this study do not depend upon illumination and occlusion variables. Our results show that facial caricature-trained neural networks outperform significantly of those

Read more

AUTHORS: Dumrong Sutapirat, Kingkarn Sookhanaphibarn and Chidchanok Lursinsap

ABSTRACT: This paper proposes a novel dewarping technique for document images based on digital cameras, from which images are warped on the curve surface of the book bound volume. Our technique transforms pixels in the curve coordinate system into the planar coordinate system. In the experiments, the test document images composed various font styles and sizes printed in common text books. The validity of our methods compares with the program’s curvature correction of Abby FineReader Professional 9.0. Our technique outperforms the other up to 49 percent for word and character precisions. Nevertheless, our technique is independent on

Read more
Page 4 of 4