The ventriloquism effect has been studied in only subject's straight ahead field. In this paper, it was measured that the difference threshold of audiovisual target and oriented position in far peripheral field. As the result, the threshold angle in peripheral field between visual and auditory target was larger than that in straight ahead field. And the oriented position was similar to auditory target position when the gap angle was not so large. This result means that, in VR systems, the audiovisual target in far peripheral field is able to have larger angular distance between visual and auditory directions than that in straight ahead field unless separating into two different directions. So it was considered to expand visible filed in Head Mounted Display by compressing vision horizontally and maintaining the sense of direction by using sound images.

抄録全体を表示

This paper describes the development of the 3DDAC mk.4 system, which is a brand new implementation in our research program. The 3DDAC is a fatigue-less stereoscopic display system that is provided with a compensating function of accommodation for binocular disparity 3D image representation. The mk.4 system also features a light-weight HMD style.

抄録全体を表示

We have studied the effect of visual stimulation presented by an 4LCD HMD which has field-of-view of 140° on postural control system from a viewpoint of the visual-vestibular interaction. In this report, the effects of visual stimulation of dynamic/static rolling computer graphics (CG) images were examined. From results of experiments, we found that, in case of dynamic image, when there is contradiction of information between visual and vestibular inputs, vestibular information is used for posture control. In case of static image, we found that, results did not show significant difference when compared with those of a closed-eye state. However, the rate of falling down was decreased by the presentation of the image by HMD. These results suggest that image presentation by HMD is helpful to standing posture control.

抄録全体を表示

In this study, we carried out an experiment to compare visual load induced by images of head mount display (HMD) with different convergence of optical system. The results showed a significant aftereffect on the accommodation response time after the subjects viewed stereoscopic 3-D images on HMD. When the convergence of optical system of HMD was mismatched with the optical distance of images, the aftereffect was remarkable. The results are discussed in terms of the mismatch of accommodation and convergence in human eyes.

抄録全体を表示

The HMD optical system composed of "Free-Form-Surface prism (FFS prism)" was presented by Canon Inc. in 1995 [1-3]. This HMD is suitable for compact HMD. We have developed a new see-through 3D HMD with high resolution, wide field of view (FOV) by improving this FFS prism technique. The new HMD with 51degrees horizontal FOV and large viewing eyebox shows clear full color image (VGA) with 920,000 pixels. In spite of the wide FOV, the thickness of this new FFS prism is very thin, 17.9mm. In this paper, we report this new HMD and "the AR^2hockey system" as an example of this HMD application.

抄録全体を表示

Conventional stereoscopic displays have inconsistent accommodation against convergence, which degrades sensation of presence. We developed a stereoscopic display applying Maxwellian optics to avoid such inconsistency by realizing large depth of focus. The display was also designed to provide large field of view (about 110 degree) with the simple optics. And this display allows an operator to observe real and virtual images in focus for wide range of depth.

抄録全体を表示

In this paper, we propose a new registration method in augmented reality (AR) system which uses information taken from a magnetic sensor and a camera attached on a HMD. In this method, the error of the magnetic sensor is corrected by tracking a landmark on the captured image. Precise registration is realized in a wide field by locating several landmarks in the scene and identifying them by using data from the magnetic sensor. Some experimental results are shown for video see-through and optical see-through AR systems.

抄録全体を表示

A design support system has been developed for constructing VR-based training environment for machine maintenance work without any expertise and programming effort on VR. Using the developed system, users can easily construct various training environments under GUI environment. It was verified through some experiments that the developed system could reduce the working hours remarkably and that novice users who had no prior knowledge on the system could construct a training environment successfully after a few hours of tutorial on the system. In this paper, the basic concept of the system, the way to model the state transition of objects by Petri net and experimental results are described.

抄録全体を表示

This paper presents a haptic simulator of the 2-dimensional peg-in-hole assembly operation in the virtual environments, considering the dynamics and the surface friction characteristics of the virtual objects. An operator can operate a virtual peg by inserting his fingers into rings at the tip of link mechanisms which display the operating feeling to the operator. By using this simulator, we can measure the force between fingers and the peg, the contact states between the peg and hole, and the position of the peg during the operation. To examin the validity of the simulator, comparison between the peg-in-hole task using the simulator and that in the real world has been done. It is hoped that this simulator will help in gathering data of human skill in assembly tasks and analyzing the human skill.

抄録全体を表示

A machine-maintenance training system was developed on an internet-based multi-user distributed virtual environment where users disassemble and assemble a pump in nuclear power plants. The developed training system has two training modes, i.e., synchronous and asynchronous modes, and it is equipped with the multi-modal communication function to provide more effective training environment. The system architecture and the functions of the system for both training modes are described in details, followed by the experimental procedure of the system evaluation. It was confirmed that the users would feel a sense of reality in the training course by the developed training system.

抄録全体を表示

This paper implements improvement of the difference between the presented depth of the virtual object and the perceived one in stereoscopic display with a large-sized screen. Spatial distribution of the depth difference is investigated to derive a suitable correction function, which adjusts pupil distance (PD) to remove the depth difference. The amount of adjustment is defined by a simple linear function depending on three variables presented depth, presented direction, and screen depth. Coefficients of them are fixed by a small number of samples of depth difference for each subject. An optical sensor is used for position sensing with high precision. As a result, the depth difference has decreased to 2.5mm on average.

抄録全体を表示

To increase the reality feeling, most researches tried to increase presentation information approaching real world. In this paper we propose a new approach of this problem. Human beings understand the real world combining many information of it. We presume the cognitive process using a human cognitive model. Then we improve the method of presenting information on the virtual world to increase the reality feeling. We will show an applying example of this approach on pseudo three dimensional display problem.

抄録全体を表示