Learn More
We propose a new scheme for practical vision systems which are simple in structure, directly and adaptively trainable for various purposes. The feature extraction consists of two stages: At first, general and primitive features which are shift-invariant and additive are extracted over the retinal image plane. Then those features are linearly combined on the(More)
This paper describes a data compression system using Evolvable Hardware (EHW) for digital color electrophotographic (EP) printers. EP printing is an important technology within digital printing, which is currently having a significant impact on the printing and publishing industry. Although, it requires data-compression to reduce the cost for transferring(More)
Our Jijo-2 robot, whose purpose is to provide office services, such as answering queries about people's location, route guidance , and delivery tasks, is expected to conduct natural spoken conversation with the office dwellers. This paper describes dialogue technologies implemented on our Jijo-2 office robot , i.e. noise-free voice acquisition system by a(More)
This paper proposes a scale invariant face detection method which combines higher-order local autocorrela-tion (HLAC) features extracted from a log-polar transformed image with Linear Discriminant Analysis for \face" and \not face" classiication. Since HLAC features of log-polar image are sensitive to shifts of a face, we utilize this property and develop a(More)
An experimental multimodal disputation system , Mr.Bengo, is a knowledge based system with multimodal user interfaces such as face recognition, face generation, speech recognition , speech generation and a WWW browser. Mr. Bengo deals with three agents-a prosecution , an defense attorney and the judge. The prosecution and the attorney dispute about the(More)
Support Vector Machines (SVMs), though accurate, are still difficult to solve large-scale applications, due to the computational and storage requirement. To relieve this problem, we propose RANSAC-SVM method, which trains a number of small SVMs for randomly selected subsets of training set, while tuning their parameters to fit SVMs to whole training set.(More)
Maximum likelihood thresholding methods are presented on the basis of population mixture models. It turns out that the standard thresholding proposed by Otsu, which is based on a discriminant criterion and also minimizes the mean square errors between the original image and the resultant binary image, is equivalent to the maximization of the likelihood of(More)
N owadays, robots are moving off of factory production lines and into our everyday lives. 1 Unlike stationary and pre-engineered factory buildings, an everyday environment , such as an office, museum, hospital, or home, is an open and dynamic place where robots and humans can coexist and cooperate. Hence, instead of the capabilities of precise motion,(More)