Hiroya Hashimoto

A low-cost real-time stereo vision system for looking at people

Real-time computer vision systems that observe the actions of people have received increasing attention as they are important for a broad range of applications. Examples are intelligent man-machine interfaces or intelligent rooms that have the capability to understand the actions of their inhabitants. In this paper we present the design of a stereo vision system for the precise and robust real time tracking of human hands and heads that unlike previous systems uses low-cost industry standard components. The system uses color and motion information to find the human. Using previously calibrated cameras the 3D positions of hands and head are calculated. Because of its client-server architecture the system is scalable to an arbitrary number of cameras and application processes. Experimental results for the performance of the system are given.

Intelligent space: a commentary on research trends

The intelligent space is a space where we can easily interact with computers and robots, and get useful service from them. In such a space, our life could be more comfortable and we can get more satisfaction. In this paper, the technologies to achieve intelligent space are introduced and problems in using these technologies are discussed. The authors introduce existing research relating to intelligent space. Finally, he concludes this paper with some argument about the future of intelligent space.

Visualization interface for AFM-based nano-manipulation

In this paper, ideal interface characteristics for nano scale object manipulation using an atomic force microscope (AFM) is discussed. Nano manipulation technology using a scanning probe microscope such as an AFM or a scanning tunneling microscope (STM) has been introduced recently. Nano world physics is different from the macro world such that the gravity can be ignored while the influences of sticking forces such as the capillary or the van der Waals forces are dominant. Furthermore, precise tip position control in the z direction is important in order to not breaking the tip or the samples. Depending on the experimental results, besides graphical displays, the visualization of the force value is important to control the z position.

Representing fundamental frequency contours generated by HMM-based speech synthesis using generation process model

Frame-by-frame representation is not appropriate for prosodic features, which are tightly related to speech units spreading a wide time span, such as words, phrases and so on. This causes an inherit problem in fundamental frequency (F0) contour generation by HMM-based speech synthesis. A method is developed to modify F0 contours in the framework of a generation process model by referring to linguistic information of input text (word boundary and accent type). It takes F0 variances obtained through HMM-based speech synthesis into account during the process. Through a listening experiment on synthetic speech, the method is proved to generate better quality as compared to the HMM-based speech synthesis on average. Since the generation process model can clearly relate its commands and linguistic (and para-/non- linguistic) information, the method has an additional advantage; changing speech styles, and /or adding further information (such as emphasis) can be easily done through manipulating the commands.

Use of fundamental frequencies shaped by generation process model for HMM-based speech synthesis

Generation process model of fundamental frequency (F0) contours is known to represent global movements of F0s keeping a clear relation with linguistic information of utterances. While HMM-based speech synthesis can generate a good quality of speech, problems, which arise from frame-by-frame processing, are pointed out. These problems are expected to be solved by incorporating the model constraints. A method is developed to use F0 contours approximated by the model for HMM training instead of observed F0 contours. A clear improvement in the quality of synthetic speech is shown through listening experiments. In the method, fragments of F0 contours not represented by the model (F0 residuals) are ignored. A scheme is further introduced to cope with the issue; F0 residuals are also included in the training and synthesis processes of HMM-based speech synthesis, and the generated F0 residuals are added to the model-based F0s before the waveform generation. The model constraint has another merit; relations between generated F0 contours and texts are clear, and it is possible to add linguistic information such as emphasis to synthetic speech, or to change speaking styles through manipulating F0s in the F0 model framework. Several experimental results supporting the advantages of the method are shown.

A low-cost, high-speed, calibrated 3D vision system for tracking humans-a component of the intelligent mechatronics network

In this paper the focus is on the second main channel of information, human pose and gesture. We present an architecture for a real-time, high precision stereo vision system to find humans, track the movement of their heads and hands and offer this data to other devices that are connected to the local area network. The system uses industry standard hardware and is capable of tracking several humans with more then 20 frames per second. It integrates into TCP/IP based PC networks and is scalable to more then two cameras. The concept is to create an area that is able to perceive what is happening in it, build a model of itself and reason about the actions of its inhabitants. Adaptive background separation is used to separate the shape of the human. Color histograms are used to identify skin areas. The detected pixels are clustered and the clusters are described by their mean and covariance matrix. Experimental results show that the system can track human hands and heads in 3D with a precision better then 10 cm. It has successfully been used to generate maps for robots from the movements of humans.