3D pointing system is required not only for measuring merely the position in 3D space but also giving information and instructions to the information system. Various types of pointing devices have been developed and used; they are fundamentally based on the position detecting function. The authors invented a new type of 3 D pointing system composed of an elastic string and three non stretchable strings with tension sensors; then proposed a calibration method for the system. They proved the functions of the device and the effectiveness of the calibration methodby manifacturing trial device based on the proposed principle.

抄録全体を表示

The Shared Mixed-Reality Space is a space where multiple people share a mixed reality world and communicate with each other. We handle not only virtual objects but also virtualized real objects in the space, and share important information even over different spaces or time. In this paper, we propose a new method for sharing attentions among people by detecting and capturing typical situation which a person is giving important explanation on an object, and by giving video clips to other people. This paper presents the basic idea, the multiple sensors simultaneously tracking important objects, and the annotation capturing mechanism.

抄録全体を表示

The measurement of a dexterous manipulation of the finger is researched. And, the measurement of the finger operation is important for a smooth information input in VR. Therefore, the measurement device of the glove type is used. However, the glove type hinders a dexterous manipulation. Therefore, the method that measures from the fingernail is researched. However, the research of the past was not able to distinguish the contact force and the bend of the finger. In this research, we use ICA(Independent Component Analysis) to distinguish the bending of the contact force and the finger. As a result, we measured bending and the contact force vector with three sensors.

抄録全体を表示

In this paper, we propose and present a new capturing system to capture and transmit video-based avatar in real time. To communicate with users from different area by using immersive projection displays, such as CABIN, we should present the user's image from various directions according to his/her relative position. However, in ordinal motion capturing system, we can't obtain photo-realistic images and detailed geometric models of human. So, we constructed surround-capturing system, which can obtain 18directional image sequences of user in blue-back booth from mounted 18 digital video cameras. By using silhouette-base recognition, we can construct 3D geometric model of remote-user in real time from the obtained image sequences. And we translate and display video avatar of the remote user in real time according to the user's position.

抄録全体を表示

This paper describes development of a food simulator. A taste of food arises from mixture of auditory, chemical, and force sensation. This food simulator is a haptic interface that can generate a force to user's teeth when the user bites a food. It consists of a film-like force sensor and one degree-of-freedom manipulator. The food simulator generates a food texture according to the force profile captured from a user's biting force of a real food. Then the profile of the biting force is replayed by controlling the reaction force of the device.

抄録全体を表示

On-demand custom manufacturing would solve discomfort problems of mass produced clothes, eye wear and footwear. With this idea, a customer's body is scanned by a 3D body scanner and the scanned point cloud is modeled into a normalized polygon model based on anatomical landmarks. Subsequently, the customer selects the several candidate products, their shapes and patterns are immediately modified to fit into his/her digital body, and virtual images of the modified candidate products on his/her digital body are displayed. The customer selects the most favorite one from candidates. He/she can also change the style, color and material of the selected candidate. To realize these ideas, we have developed a low-cost and compact foot scanner and methods for homologous human body modeling and calculating representative 3D body forms of a population. Other techniques necessary for on-demand custom service such as draping simulation, searching technologies based on customer's KANSEI are discussed.

抄録全体を表示

In this paper, we present a new approach to computer-aided training in Medicine based on the visualisation of simulated clinical conditions using a virtual patient. Our system uses a knowledge-based model of cardiac physiology to generate realistic emergency situations (shock states). The overall results of the simulations are presented through a realistic 3D virtual patient whose behaviour and appearance are consistent with the clinical data generated. After introducing the principles behind the simulation and the system's architecture, we describe the process through which physiological parameters are mapped to the visual appearance of the 3D patient and discuss representative results from the system.

抄録全体を表示

The topic of this paper is interaction/information environments in our daily life space. When designing these systems, it is thought to be important that they match where they are placed and that they have congruous interfaces. To realize the concept above, the authors propose the method "Enhance objects based on their own metaphor". Based on this method we implemented two i-mirror (intelligent/informative mirror) systems (each has different optical system). Basically, the i-mirror systems imitate a mirror using special optical systems, and the applications are based on virtualizing and enhancing features of a mirror. Four mirror-metaphor-based (interactive) applications (a mirror with gain, a mirror with memory, a look-younger/older mirror, and an intelligent lounge with a mirror searchable past scenes) were implemented on the i-mirror systems. We found the interfaces brought from the concept effective for congruous interaction, from users' reaction.

抄録全体を表示

The sound field reproduction technique based on the Huygens's principle is important to realize auditory virtual reality. However, it is necessary to transmit many multi-channel signals. In this paper, a new spatial coding technique for the multi-channel signals is proposed to reduce transmission bit rate. Source signals are extracted by convolving the inverse transfer function calulated from a room transfer function to the channel signals at an encoded side. At a decoded side, channel signals are reconstructed by convolving the room transfer function to the extracted source signals. As a result, the size of data to be transmitted is reduced from the number of channels to that of sound sources. A coding experiment was performed for a real sound field. It was confirmed by subjective assessment that the perceptual distortion of the coding had little effect on the perceptual quality of sound field even if the source signals are not completely extracted.

抄録全体を表示

In the field of mixed reality, virtualization of real outdoor scenes is widely investigated for a number of applications such as telepresence and walk-through systems. An omnidirectional multi-camera system is more useful for virtualizing outdoor scenes than a single camera system, because the multi-camera system can easily capture high-resolution omnidirectional images. However, the exact calibration of the system is necessary to virtualize the real world accurately. In this paper, we describe a method for generating a high-resolution panoramic movie by calibrating an omnidirectional multicamera system geometrically and photometrically. In experiments, a spherical panoramic video has been successfully computed to show the validity of the proposed calibration method. Finally, we show a prototype of telepresence system using an immersive display.

抄録全体を表示