This study investigated what effect physical constraints have on the interpretation of demonstrative pronouns when a user navigates a robot. For this investigation, a robot navigation environment called Spondia-II was develope, and an experiment conducted. It is known that the interpretation of demonstrative pronouns requires information about not only the situation (or context) but also the speaker's viewpoint during a dialogue. The results of the experiment suggest that physical constraints do affect the user's viewpoint, especially when a user utters a demonstrative pronoun while navigating the robot. In actual fact, the user alters the use of demonstrative pronouns according to the change in the user's viewpoint. It is also suggested that the user and the robot share the same viewpoint during the physical interaction.

By sharing data regarding the sensations experienced by individuals, as well as by sharing their knowledge, we are readily able to communicate with each other, and there are possibilities to further evolve this communication method. The different sensations experienced during voluntary walking and enforced walking give us different feelings. Also, the number of individuals involved can create a different feeling when walking. Networked computer-assisted walking can support and enhance these different experiences. In this paper, we introduce another walking style, the shared power-assisted voluntary walk, which is realized by a prototype networked locomotion system. This system can be used in tele-rehabilitation, which allows remote patients to share the sensation of walking. Also, it can be used to teach a group of patients rehabilitative walking. We developed two locomotion interfaces and connected them via a network. We developed enforced and semi-voluntary walking training systems using the shared walk environment and evaluated them with a series of experiments. operation integration process. In the second algorithm, thanks to deferred broadcast of operations to other sites, this process becomes even more simplified.

Describes a series of stages in the development of a new virtual locomotion device that has been designed to enhance remote inter-personal communications. The latest system, called GSS (Ground Surface Simulator), inherits the features of two previous locomotion interfaces, viz. ATLAS (ATR Locomotion interface for Active Self-motion) and ALF (ALive Floor). GSS also incorporates two different features not found in ordinary treadmills: a movable belt and an active belt-speed controller. We built an initial prototype of GSS and developed a method that presents bumpy surfaces that are free from the mechanical limitations inherent in prior designs. Experimental results showed that a subject could distinguish a 1% difference in the virtual slope on the GSS.

This paper describes an investigation ofthe generation of surrealistic spaces that aim to achieve a new communications environment in which people can mutually con'vey their thoughts, mental images, and passions.Passion spaces here can be compared with conceptual spaces in the analogy of "logos versus pathos."Our &cus was on the phenomenon of synesthesia, and the results shown are the tests of color image creation involving "tanka" (31~syllable Japanese poems), aiming at the generation of spaces where some passions are reffected rather than concrete scenes imagined Corn word meanings.

The paper describes an investigation of the generation of surrealistic spaces that aim to achieve a new communications environment in which people can mutually convey their thoughts, mental images, and passions. Passion images here can be compared with conceptual images in the analogy of "logos versus pathos". Our focus was on the phenomenon of synesthesia, and the results shown are of the testing of color image creation involving "tanka" (i.e., 31-syllable Japanese poems), aiming at the generation of images where some passions are reflected rather than concrete scenes imagined from word meanings. In addition, the paper proposes an application that judges the similarity of pieces of music, showing the possibility of making instant, subjective judgements without the need for a performance.

This paper describes an investigation of the generation of surrealistic spaces that aim to achieve a new communications environment in which people can mutually convey their thoughts, mental images, and emotions more effectively. Emotion spaces here can be compared with conceptual spaces in the analogy of "logos versus pathos." Our focus was on the phenomenon of synesthesia, and the results shown are the tests of color image creation involving "tanka" (31-syllable Japanese poems), aiming at the generation of spaces where some emotions are reflected rather than concrete scenes imagined from word meanings. In addition, this paper proposes an application that judges the similarity of pieces of music, showing the possibility of making instant, subjective judgements without need for a performance.

We report preliminary results on emotion recognition by machine from joint audiovisual input of facial video and speech. The results show potential advantages in using both modalities over either modality alone. The recognition rate for audio alone is about 75% and video alone about 70%. Using audiovisual data we achieved 97% without increasing the number of features. The improvement in performance is accredited to the complementary property between the two modalities. A possible application is in natural human-computer interfaces.

We developed a mechanism in which the behavior of actors in computer-generated virtual space is determined in response their interaction with each other. We call this mechanism the Autonomous Interactive Reaction model (AIR model). The AIR model is a simplified mechanism for determining human-like behavior expressed in terms of parameters for emotion and personality, and an actor's behavior is determined using these parameters. The AIR model adjusts these parameters in response to the communication between actors, so each actor's behavior is determined as a result of his interaction with other actors.

Palmtop displays have been extensively studied, but most of them simply refocus information in the real or virtual world.The palmtop display for dextrous manipulation (PDDM) proposed in this paper allows the users to manipulate a remote object as if they were holding it in their hands.The PDDM system has a small LCD, a 3D mouse and a mechanical linkage (force display).When the user locks onto an object in the center of the palmtop display, s/he can manipulate the object through motion input on the palmtop display with haptic sensation.In this paper, the features of the PDDM with haptic sensation are described, then four operating methods and the haptic representation methods for a trial model are proposed and evaluated.(see Video Figure in the CH196 Video Program)

We describe a method to detect hand position, posture and finger bendings using multiple camera images. Stable detection can be achieved by using skeleton images, and this is confirmed through experiments. This system can be used as a user interface device in a virtual environment, replacing glove-type devices and overcoming most of the disadvantages of contact-type devices. Future work includes verification of the method's availability as a gesture-based man-machine interface system.

An experimental Japanese speech communication aid using speech synthesis, and its operation principle, are described. The results of a short evaluation are presented. The system, VOICE-AID, can be used to carry on a telephone conversation from anywhere over telephones in homes, offices, or public telephone booths. Neither special equipment nor an adaptor is needed to use the system.>

The authors describe a data syntax conversion method from CEPT (European standard) to Captain PLPS (Japanese standard) to permit interconnections between videotex services. They discuss the conversion results when an overseas videotex center, in particular, the Prestel (alpha-mosaic version in CEPT standard) center in the United Kingdom, is accessed from a Captain terminal using the described method. By the conversion method presented, the Captain terminal can access CEPT videotex centers as a CEPT terminal. There are two essential features. The first is that text and graphics data are converted into photographic information on Captain terminals to absorb differences of character size and graphic display. The second is that a color replacement method based on the major colors in a coloring unit is used to solve the problem of differences of coloring units between the two videotex services.>