We propose a novel approach to face identification and verification based on the Error Correcting Output Coding (ECOC) classifier design concept. In the training phase the client s...
Josef Kittler, Reza Ghaderi, Terry Windeatt, Jiri ...
It has frequently been observed that most of the world’s data lies outside database systems. The reason is that database systems focus on structured data, leaving the unstructur...
Alon Y. Halevy, Oren Etzioni, AnHai Doan, Zachary ...
Abstract. We propose a vector representation approach to contour estimation from noisy data. Images are modeled as random elds composed of a set of homogeneous regions contours (bo...
In various applications such as data cleansing, being able to retrieve categorical or numerical attributes based on notions of approximate match (e.g., edit distance, numerical di...
Liang Jin, Nick Koudas, Chen Li, Anthony K. H. Tun...
Background: Agglomerative hierarchical clustering (AHC) is a common unsupervised data analysis technique used in several biological applications. Standard AHC methods require that...