-
Article type: Cover
Pages
Cover1-
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Index
Pages
Toc1-
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Hiroo IWATA, Wataru HASHIMOTO
Article type: Article
Pages
1-6
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper presents design method of rear-projection based dome screen. Spherical screen is ideal for spacial immersive display, because distance between the viewer and the screen is constant. Spacial immersive display must use rear-projection screen. If the screen is front-projection, the viewer obstruct the projected image. This paper discusses issues in design of rear-projection dome screen such as choice of projectors, determination of radius of the screen, and distortion correction.
View full abstract
-
Kinya FUJITA, Takuro KAWACHI
Article type: Article
Pages
7-10
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
A glove-shaped simple force feedback system to display hardness of virtual object was developped. The principal design was to apply reaction force on fingertips of thumb and index finger by using two wires driven by position-servo-controlled DC-motors. The joint torque to PIP and MP joints was applied with single wire but at appropriate torque distribution ratio by optimizing the structure of the glove system. The performance of the system was evaluated through distinguish test in 12 volunteer subjects. The average cognitive rate at four levels of hardness was 73%.
View full abstract
-
Shogo FUKUSHIMA, Hidekazu YOSHIKAWA
Article type: Article
Pages
11-16
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
A new-head mounted display, Eye-Sensing Head Mounted Display (ES-HMD), has been developed which can not only provide the user with virtual environment through visual and auditory stimulus conventionally but also monitor the user's both eyes and calculate ocular information such as pupil size, eyeblink or eye gaze. The information can be expected to be utilized for analysis of human internal state such as mental workload, cognitive state or longtime fatigue which affects such ocular information while watching display image. In this paper, the results of the developed HMD are introduced with respect to basic functions of ES-HMD, followed by the discussion on the basic experiment of eye tracking by using the ES-HMD.
View full abstract
-
Junji FURUSHO, Masamichi SAKAGUCHI, Masafumi SENDA
Article type: Article
Pages
17-20
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed ER actuators with low inertia. These actuators have a good performance for force display devices in virtual reality. Then we made a force display system which is connected with CG. Using this system, we conducted force display experiments. Finally, several applications of ER actuators to virtual reality are proposed.
View full abstract
-
Hiroki MORI, Akira UTSUMI, Jun OHYA, Masahiko YACHIDA
Article type: Article
Pages
21-26
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We describe a system to estimate position and orientation of a human body. In such a system, deformation and occulusion of the human body often cause difficulties in the estimation. Our system employed multiple-camera based method to avoid these difficulty. Skeleton images detected from multiple cameras are utilized for stable estimation of both posltion and orientation. Furthermore, according to the estimation result, the best camera for the body side detection is selected. We confirmed the stability of those through experiments.
View full abstract
-
Yutaka KUNITA, Masahiko Inami, Taro MAEDA, Susumu TACHI
Article type: Article
Pages
27-32
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose a new method how to construct three dimensional figure of humans and objects in a virtual environment. The process of this method is so simple-to select, scale and reconstruct scanning lines of several cameras rolling around an real object-that it can be realized by simple hardware. By applying this method to the tele-conference system that we previously reported, plural humans at the distance can exchange photo-realistic images of themselves with one another in real time.
View full abstract
-
Manabu NAKAMURA, Hiroyuki MURATA, Masahito OGATA, Kazunari MORIMOTO, T ...
Article type: Article
Pages
33-38
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The aim of this paper is to verify whether our adjustment method of depth distance is useful to create objects in a virtual space. In this experiment real and virtual objects presented in a real and a virtual spaces respectively were pointed by the index finger of the right hand. After watching a virtual object subjects calculate the distance from their position to the object, and their hand movement starts to the object. In order to point the object exactly subjects have to coordinate human vision and perception in a fusion space of a real and a virtual spaces. In the experiment we measured the pointing accuracy of virtual objects presented with or without use of depth adjustment through our experimental data of depth perception. Furthermore the pointing accuracy of real objects presented in the real space was measured. Results show that the pointing accuracy in a fusion space highly increased by use of the depth adjustment method, however, was slightly lower than that in the real space.
View full abstract
-
Michitaka HIROSE, Shinjiro WATANABE, Takaaki ENDO
Article type: Article
Pages
39-44
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
To generate 3-D virtual spaces, the quality of images has become an important issue. In order to generate more realistic virtual spaces, we are developping a new image capturing and generating system. This system includes an automobile equipped with multipul CCD cameras for image capturing and GPS and Gyro for position sensoring. This onboard system produces largescale image data base indexed by positional information, from which images corresponding to the user's location should be able to be displayed properly. At the end of this paper, we also refer to the problem about this methodology in synthesizing a smooth image motion.
View full abstract
-
Hiroshi Iezaki, Yasushi Ikei
Article type: Article
Pages
45-51
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes the design scheme that satisfies the technical requirements which arise when we build the real-time distributed virtual environment that includes multiple simulation computers connected by the Internet. We argued the methods in which the trajectories of objects around the collision were presented continuously and with less errors in the shared space. After the classes that defined the position and the behavior of objects were classified, the object trajectories were examined by the experiments in terms of four simulation models. The trajectory data were shown quantitatively, and the characteristics of the models were discussed from the points of the event consistency, the real-time interactivity, and the behavioral presence.
View full abstract
-
Ken-ichi Kameyama
Article type: Article
Pages
53-56
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes a system called styling CAD for an industrial design in which a stylist can intuitively make solid model data and can collaborate with other designers. The key components of its hardware are a special input device with which a stylist can directly manipulate a surface of a virtual object as he/she manipulates a clay model. The software consists of a graphical viewer for displaying an object in 3-D, a data translator with a solid modeler for producing solid model data, and a communication interface for transferring data to other systems for engineering and manufacturing. Using this technology, a powerful design environment which supports a design process from beginning to end can be realized. A preliminary prototype environment for a non-spherical lens design is mentioned as an example.
View full abstract
-
Tetzuo Tezuka, Ryuji Inoue, Hidekazu Yoshikawa
Article type: Article
Pages
57-66
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We have developed a teleoperation system for robot-arm manipulation based on virtual environment. The virtual environment for manipulating the robot arm is constructed by measuring the 3D positions of the objects around the robot arm. The 3D position is estimated by using two-planes method based on camera images captured by the CCD camera on the arm. In this study, the precision of this 3D-measurement is evaluated through experiments and also by using a theoretical model. This system is applied to the experiment of Peg-in-hole practice.
View full abstract
-
Naoyasu HIRAO, Futomi SHIMONO, Masahiro KIMURA, Mieko OHSUGA
Article type: Article
Pages
67-72
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We refer to the preliminary results on the assessment of mental stress during a simulated cooperative work using physiological data, subjective one and work records. We developed a simulated crane task using virtual reality techniques. Four pairs of male subjects participated in the experiment. Their physiological and subjective data and work records were collected and analyzed. Using stress detection method triggered by physiological data, we obtained effective in formation to improve the cooperative Work. The entrainment phenomena between the two subjects'physiological data in the cooperative work were not ovserved clearly.
View full abstract
-
Michiya Yamamoto, Tetsuo Tezuka, Hidekazu Yoshikawa
Article type: Article
Pages
73-80
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Recently, it has been a serious problem to maintain and improve the reliability of the engineering system, which tends to become more complex than before with the advance of technology. However, the training education to maintenance personnel becomes more important for improving their skill, but the conventional practice using real size machine models can only afford limited opportunity to the trainees. In this study, we have developed a Machine-Maintenance Training System with networked virtual environment for multiple trainees. The configuration and the performance of the system are described in this paper.
View full abstract
-
Kouichi NAKANO, Yasuo WATANABE
Article type: Article
Pages
81-88
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper presents a CAD system for traditional craft which a designer can deform the object in a virtual space by "push"or"pull"operations as he touches it directly using Cyber Glove. The method for non-contact measuring the 3D shape was also developed for the purpose of inputing the shape of traditional craft works in high accuracy. The method can improve the accuracy by interpolating the missing data caused by self-occlusion. The simulation of external appearance for a new design of traditional craft works became possible by developing a CAD system with VR.
View full abstract
-
Takashi UENO, Yasuhito YAMAMOTO, Takuji KOBAYASHI, Genta GOTO, Yuji NA ...
Article type: Article
Pages
89-92
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We have been developing the Hyper Hospital, a network based VR mediated medical care system. The Hyper Hospital is composed of two seamlessly integrated environments, that is, the virtual and the real worlds. Of them, its virtual environment expands the conventional medical care system using the virtual reality technology as a principal human interface and a collaboration tool. In the present study, an attempt to extend the Hyper Hospital system to a satellite communication based multi-media network using the Internet protocol is reported. Using the WWW as the human interface, transmission rate and other performances were measured with various data, including VRML pages and realistic medical image data. The satellite network was shown to be suitable to the Hyper Hospital network, and a plausible application of the network was discussed.
View full abstract
-
Nobuyuki OZAKI, Kaoru TAKADA
Article type: Article
Pages
93-98
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper presents two practical applications for virtual shopping. Virtual reality technology, especially Image Based Rendering technique, enables us to exhibit products in sophisticated way. With a concert ticket reservation system, one can confirm the view of the stage from a particular seat you want to buy. These images are synthesized from the captured images and are not created from the computer graphics, which is less realistic. With a catalog shopping system which uses captured images to show products, one can observe them in a stereoscopic way and change angles of the viewpoint interactively. These systems are implemented into the KIOSK type terminal, which is supposed to be placed where people gather. One can purchase with IC-card type electronic cash. These two systems will be exhibited at Toshiba Science Museum at Kawasaki, Kanagawa.
View full abstract
-
Atsushi KUBODERA, Takashi TAKEDA
Article type: Article
Pages
99-104
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper discusses game software designed using an "Action Interface"-hand position recognition interface-proposed in a previous paper. It is proposed that this software be used for patient rehabilitation purposes as well as for physiological examinations such as testing patient's hand to brain coordination.
View full abstract
-
Hiroaki SHIGEMASU, Takao SATO
Article type: Article
Pages
105-112
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, the effect of combination of stereopsis and structure from motion was investigated. We discussed whether the addition of motion information overcomes the stereo distance scaling problem. We manipulated the richness of monocular depth cues and examined the effect. Our results reveal that when stereo cue and motion cue are integrated the richness of information of monocular depth cues effects the perception of depth and stronger depth cue determined the integrated depth perception.
View full abstract
-
Takao Sato, Takashi Okubo, Hiroaki Shigemasu
Article type: Article
Pages
113-120
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
When the two pictures of a stereo pair are interchanged, binocular disparity is completely reversed. Thus we should perceive reversed depth, if perception is solely determined by disparity. However, we tend to perceive normal depth, i.e. a normal convex face instead of concave face (reversed disparity illusion). This interesting phenomenon was studied systematically through several psychological experiments with photographs of human face. The illusion weakened when disparity was exaggerated, or when the familiarity of the stimulus was decreased by presenting negative or up-side-down photographs. These results indicate that the illusion occurs through complex dynamics among different depth cue systems.
View full abstract
-
Jun'ichiro Seyama, Takao Sato
Article type: Article
Pages
121-129
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
To estimate shape from shading, the relationship between image intensity and surface orientation, reflectance map, is required. In this study, subjects estimated the slant perceived from images of cylinders. The reflectance map used by the visual system was estimated from the relationship between the perceived slant and stimulus image intensity. Results suggest that the visual system can modify the assumptions undertying the reflectance map, surface reflectance properties, illuminant intensity, and ambient light.
View full abstract
-
Naoyuki Matsuzaki, Takao Sato
Article type: Article
Pages
131-136
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Johansson (1973) showed that human movements such as walking can be recognized from a small number of dots fixed at various parts of the body (biological motion). We investigate whether people can also recognize facial expressions of emotion from movement of a small number of dots and the roll of motion information on recognition of facial expressions. Our results reveal that motion of as few as 18 dots attached to appropriate locations on a human face provides sufficient information to recognize facial expressions.
View full abstract
-
Article type: Appendix
Pages
App1-
Published: June 03, 1997
Released on J-STAGE: June 23, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS