We propose a tactile touch pannel in controlling the displacement of the thin film by the shape memory alloys.

抄録全体を表示

In this paper, we propose an estimation method of window area ratio in order to utilize in a computer simulation of radio wave propagation. For high-rise buildings we employ bird's eye view images to get clear wall images with few occlusions, and develop a novel method to detect corners of buildings for regularization of the wall. Our method achieves accurate estimation for real building images.

抄録全体を表示

We proposed the rotated character recognition method using Parametric eigen-space method . In this report, we present a real-time recognition system for rotated character images acquired from a web camera.

抄録全体を表示

Regarding the ear biometrics system the authors have proposed, robustness against pose variations are thoroughly examined.

抄録全体を表示

The authors study a metric to measure the perceptual quality of coded pictures based on parametric No Reference framework. The proposed objective quality index employs three kind of image features; quantiser-scale, blockiness and flickerness. Performance evaluation shows that the proposed method has a correlation coefficient of 0.867 and outperforms the PSNR metric.

抄録全体を表示

This paper presents a method for dance evaluation based on principal component analysis of motion pictures. The method allows us to evaluate rhythm factor of dance without complicated preprocessing. Experimental results show that it is possible to specify rhythm for a static background of images.

抄録全体を表示

We studied the changes in dynamic visual acuity (DVA) performance by looking at differences in conditions, and successfully verified the improvement of DVA performances in dominant eye viewing compared to non-dominant eye viewing when looking at Landolt C, but not when looking at E chart.

抄録全体を表示

For eye gaze detection, we detect the pupil and the corneal reflection of the light sources, using the two light sources and the image difference method. However, the detections under highly illuminated environments are difficult especially when the head moves. We improved the detection method by using a high-speed camera.

抄録全体を表示

CAVE-like multi-screen display often leads to a strong sense of self-motion when an observer views a moving pattern on a display. In this study, we evaluated the strength of self-motion by measuring subject's body sway by using a balance board accessory for the video game console.

抄録全体を表示

This paper introduces the proposed driving ability measurement simulator and methodology, providing sample results for driving ability (visual and cognitive functions) among the elderly.

抄録全体を表示

In this paper, it was examined whether the immediate pre-driving training (brain activation training) was effective in improving the cognitive functions (attention and decision-making) in driving.

抄録全体を表示

In this paper, we propose a method for detecting decline in older driver's cognitive function and screening dementia at an early stage by observing the driver's behavior and measuring their driving performances using a newly developed driving simulator.

抄録全体を表示

There has been an increase of serious traffic accidents in the past few years such as by drivers suspected of having dementia. License renewal for such people has become a big issue in the society. This study aims at developing the automatic test system for dementia screening.

抄録全体を表示

We propose a technique for human behavior detection from a surveillance video. Our proposed system identifies 'ObjectPut', 'Pointing', and 'CellToEar' events based on the bag-of-features approach. We used a fixed-dimensional descriptor that was extracted from a key-point trajectory as a feature. Our system achieved a relatively robust performance.

抄録全体を表示

Criminal acts in taxis and vehicle theft occur frequently, consideration of systems relating to vehicle security is an important issue. The purpose of this research is to develop a system that can detect suspicious behavior of passengers and vehicle theft, and automatically detect and record criminals.

抄録全体を表示

This paper describes a method for detecting if a driver can drive safely by measuring pupillary light reflex to active photic stimulus before he/she starts driving. We conducted some experiments on drunk and sober drivers, and confirmed that our index "63% dilation time" is effective to classify those drivers.

抄録全体を表示

This paper describes a method for detecting fatigue states caused by deoxygenated hemoglobin of blood before he/she starts driving.we done experiments on discriminate each states by use of neural network which was built by hyper spectrum camera datas.

抄録全体を表示

Our purpose is a verification of the influence that the processing level exerts on memory of contents with scent. The image picture and the image word are included in contents. It is an experiment that recollects imaging, and memorizes the scent.

抄録全体を表示

There is a symptom that is called VR intoxication in three dimensions VR space. The smell was effective in the recovery of the symptom or it researched. And, easing the symptom by the smell was researched.

抄録全体を表示

the effect of the fragrance is applied to the aromatherapy, and the application to relaxation and arousal effect can be also expected. We obtained verification results that fragrance presentation is effective in awaking the drowsy driver. Then, it has aimed to verify the effect in relieving fatigue by fragrance presentation.

抄録全体を表示

The falling asleep at the wheel is big problem that should be exterminated. Then we pay attention to the method by "Fragrance presentation" with which it doesn't interfere in the driving operation and are examining it. We tested difference of alerting duration by the kind of smell and presentation method.

抄録全体を表示

In the animation of facial expression with point clouds (PCs) from Range Images, looking for corresponding points in different PCs is a critical problem for local face motion while global face motion can be decided by ICP algorithm. The local face motion is a set of partial movements on the face. Therefore in this paper we propose a method of region segmentation and interpolation for generating interpolated facial expressions between two different facial point clouds. With the interpolated points of the pair of corresponding regions at time T, the animation of facial expression can be generated.

抄録全体を表示

This paper describes a method of reconstructing fractures of an earthenware that is lack of an object surface. An N-sided hole filling algorithm are employed to generate surface of fracture.

抄録全体を表示

The present paper describes a method for simulation of glasswork. The aim of the present study is a proposal of the glass model that the split of the shape not considered in the glass model of an existing technique is possible. This study, by examining the split of the glass model using the proposed method, and verified its effectiveness.

抄録全体を表示

We developed a virtual studio system that consists of simple instruments only. Introducing a chroma-key synthesis function into TVML Player, we achieved desired function and confirmed its performance.

抄録全体を表示

This paper describes how to determine haptic device orientation to construct a three-finger grasp system. According to the human hand construction, orientation of the three haptic devices are derived by our calculation model. As the result, range of proxy movement of haptic devices will be optimized.

抄録全体を表示

We are researching MR-PreViz system for filmmaking. This paper proposes a toolkit for designing action scene and recording camera-work.

抄録全体を表示

We are researching MR-PreViz system, which utilizes mixed reality technology for filmmaking. In this paper, we introduce a scalable MR compositing system and a method for on-site real-time 3D match move for MR-PreViz.

抄録全体を表示

In this study, A Retro-Reflection type large screen autostereoscopic display providing two or more viewers is proposed.We made a prototype system and confirmed the availability of the proposal system.

抄録全体を表示

We report a large-scaled phase electroholography display system for 3D reconstruction. It is based on Fresnel computer-generated hologram, and is developed by use of GPU cluster, phase-modulator-type LCD panel and the wavefront-recording method. The wavefronf-recording method is two-steps algorithm. In this paper, we implemented the second-step (diffraction calculation) onto GPU.

抄録全体を表示

A wavefront recording method is capable of calculating a digital hologram at high speed; however, it does not treat hidden-surface processing. We report digital hologram generation from a 3D object with hidden-surface processing using wavefront recording method.

抄録全体を表示

We are developing multiple video acquisition system by mobile phones. In this system, a frame synchronization is needed for obtaining three dimensional scene situation by integrate them. In this paper, we explain two types of frame synchronization. One is sound, and another is NTP. We also evaluate accuracy of synchronization methods by experiments.

抄録全体を表示

In this paper, we propose a control method for multiple robot cameras. Experimental results show that users can turn all robot cameras toward a subject and shoot the multi-view images by this proposed method. The multi-view images are useful for generation of 3D models.

抄録全体を表示

The distortion of the barrel type is caused by liquid-crystal varifocal lens for volumetric three-dimensional display. Distortion was measured and fitted to theory in liquid-crystal varifocal lens for aberration compensation. The input image is successfully transformed from the obtained distortion to compensate.

抄録全体を表示

We propose a simple viewer for both DFD(Depth-fused-3-D) display and stereoscopic display. This viewer enables us to evaluate the perceived depth of DFD display and stereoscopic display at the same time. The perceived depths of both displays were also verified experimentally.

抄録全体を表示

An appropriate binocular vision was able to be offered was verified if the individual variation generated when it watched the binocular vision was investigated, and it adjusted it very.

抄録全体を表示

We propose a crosstalk simulation algorithm for frame sequential 3D TVs. Our algorithm precisely simulates how the light from a back light reaches human eyes through a LCD panel and shutter glasses.

抄録全体を表示

We are studying on improvement of ISDB-T HDTV mobile reception technology for expanding reception area. This paper describes the performance of field experiments using the receiver that adopts 8-branch space diversity reception technology and iterative decoding technology between Viterbi decoding and Reed-Solomon decoding.

抄録全体を表示

DTTB (Digital Terrestrial Television Broadcasting) has been started in December 2003, and the receiver for DTTB is put on the market by a lot of enterprises. However, the reception performance of the receiver is different from one another. In this paper, we describe the results of the performance assessment.

抄録全体を表示

In the DTTB (Digital Terrestrial Television Broadcasting), the error correction capability of the receiver determines required C/N(carrier to noise ratio). We have improved the error correction capability of the Viterbi decoder by use of error information from RS decoder.

抄録全体を表示

This paper describes our developing P2P Live Streaming System, which improves utilization of high speed terminals. We evaluate its performance of server load reduction and stable distribution from the result of live streaming event of NHK schools chorus contest.

抄録全体を表示

We are developing Mesh P2P streaming technologies for delivering various programs over the Internet. In this paper, for the effective utilization of the narrowed band frequency utilized by the terminal and for the control of the traffic triggered among line operators, we analyze the stability in delivery when the priority of connecting peer is set.

抄録全体を表示

ARIB Technical Report TR-B29 specifies metadata of video and audio features and the transport methods in order to facilitate efficient and accurate monitoring of broadcast chains. We have developed a monitoring system using the metadata that detects faults of video and audio signals on a broadcasting chain.

抄録全体を表示

We introduce a trial of TV live-program using Twitter and USTREAM.

抄録全体を表示

We are developing a system that augments broadcasting programs with content provided over broadband networks. This paper describes an implementation of syncronization function between broadcast and IP delivered content in an embedded reciver.

抄録全体を表示

This article describes our proposed new transport scheme to enable combining multiple components on broadcast and broadband, and also shows results of verification experiments with a prototype system.

抄録全体を表示

NHK is developing an advanced broadcasting platform "Hybridcast," which combines broadcasting and a broadband communications network together to provide services that are more convenient and enjoyable for viewers. This paper describes a target architecture of the Hybridcast system and a prototyping of its basic functions.

抄録全体を表示

In order to estimate the coverage area for mobile reception of ISDB-T_<SB>, we made a field experiment and measured the relation between the Erroneous Second Ratio and the measured electrical field strength.

抄録全体を表示

We carried area 3D broadcasting live by using the terrestrial digital broadcasting..

抄録全体を表示

In order to estimate the service area of a mobile multimedia broadcasting which is scheduled to operate in the VHF-band, we measured the electric field strength of Digital Radio wave in mobile reception. This paper discusses the result of the measurement.

抄録全体を表示