This paper presents a study for creation of swimming animation using multiple view. The authors experimented to get 3D
motion data using stereo video images of underwater and demonstrate diving animation as a result.
We propose a fast rendering algorithm to create photo-realistic trees in various kinds of daylight. Utilizing a two step shadowing algorithm, the quasi-3D trees can create high quality forest scenes at a low cost. This is especially useful for the environmental assessment of landscapes.
For graphical viewing of the tertiary structures of proteins, the coordinate locations of atoms were automatically read out from protein data files, and these data were processed for this purpose. In this study, we present graphical images generated by using myoglobin and prion data.
3-dimensional CG animation has the advantage which promotes an intuitive understanding. However, The animation becomes complicate when it has much information to visualize. Consequently, the understanding by a user is barred. The contents which combined interpretational documents and 3DCG animation were created in this study.
In case of learning the motion of human body the distance-learning systems are focused on not only how to exercise, but also on how to explain the meaning of the motion. Therefore, we propose the system of learning the movement of human body with distance-learning environment.
We reported the 3D image modeling method with multiple-cameras arranged around the object. We have developed a new 3D browser for displaying the 3D image models, generating a new scene composed of CG characters and CG background. As the 3D image model data was transformed to binary data, the browser can display 3D image sequences faster than the conventional browser.
We have proposed a new block matching method for ray space interpolation using offset block matching. The proposed method has gained the advantages of block-based and pixel-based methods. Block-based ray space interpolation has been proposed to enhance the pixel-based. Our proposed method can get better interpolation result than the block-based method.
In image media, there are growing interests and requirements for realistic 3D images rather than 2D ones. Many people research on a 3D image processing field, such as building of virtual space using 3D images. The 3D image information is acquired by LASER imaging devices, Range Scanner. In most case, the available 3D image information processing technologies are much depend on imaging devices and Computer Graphics.However, current Range Scanner and Computer Graphics don't give realistic enough visual effects. There are many no acceptable errors, due to the accuracy limit of Range Scanner, and many occluded areas by trees and electric cables. Therefore the problem of occlusions is taken up and Level Set Method, which is of wide application to the boundary value problem and the initial value problem, is applied to this problem.
We present a method to repair range data in those defect regions from observed data in their neighborhoods. Our method applies the interleaved sequential updates, composed of the transportation-based inpainting and the data projection onto the original viewing direction of each range-data sample.
This paper presents a method of the image post-production utilizing range data observed with a laser radar range finder, and addresses the problems thatwe need to solve. As a preliminary experiment, we post-produces fog effects by utilizing the observed range data
The 3-dimensional(3-d) frequency [horizontal-vertical-temporal frequency] domain can be understood easily and directly by the newly proposed SZB(Spherical Zone Ball), which is the direct extension of the 2-d CZP(Circular Zone Plate) to the 3-d one. This paper also proposes the direct-observing method of the 3-d frequency characteristics.
In the multi-camera system which used the plane mirror, when get the image which photographed single viewpoint, the im-age information on the boundary line of a mirror and a mir-ror will suffer a loss, and edge will arise between the con-nected images. Then, although the image which does not have a deficit by shifting and photoing a viewpoint position is acquired in this paper, the optimal photography for con-sidering the influence on object connect from which the depth at the time of single viewpoint restrictions un-corresponding differs, and connecting is described.
This paper describes a system generating 3D human model by using 19 IEEE-1394 cameras. An object of moving human is shot with 19 cameras, and the 3D shape of the human body is estimated by a hybrid method utilizing the block matching method and the volume intersection method to generate a model. The model data is described with VRML standard.
A basic study for developing a small type remote processor of 3D image transmission employing juice-can(CanSat) is studied . In CanSat fundamental electric devices for image transmission should provide several functions compatible with severe limitation of frequency bands . Here are image transmitting technique using SSTV which enables us to make band lengths within 3.0KHz by transforming image data into audio signals is studied . In this paper , this technique is applied to a transmission of 3D images , using the hologram which enables us to record in three dimensional space as fringe patterns . As a result , excellent 3D images are reconstructed from the transmitted data of the hologram adopting SSTV process .
In this paper, we attempt to develop "Wearable Imaging System" that is capable of constantly capturing data not only from wearable video camera, but also from various kinds of sensors such as brainwave analyser, GPS, gyro sensor.In addition, the system provides interface for efficient video browsing and retrieval with data from these sensors.
Adding color to a grayscale image requires a lot of try and error. We propose a new method to colorize the grayscale image using a neural network. The neural network learns a relation between brightness and the color in a reference color image. The grayscale image is colorized through the network.
In OFDM transmission, it is difficult to improve preformance byequelization under severe frequency selective fading. In this paper, subcarrier power control of data spread OFDMapplying transmission diversity scheme is proposed. Performance of the proposed scheme is examined by computer simulationand good performance is confirmed.
In OFDM transmission, a loss of sub-carrier orthogonality due to Doppler-spread of the channel leads to inter-carrier interference (ICI).We have proposed a method to reduce the ICI caused by the Doppler spread ofthe channel.The transmission channel is modeled by a combination of multipleDoppler-shifted propagation paths, and their parameters such asattenuation, relative delay and Doppler-shift are estimated usingscattered pilot symbols.The ICI is canceled by multiplying the inverse matrix of theestimated channel matrix with the received OFDM symbol vector.However, the ICI canceller causes enhancement of noise component,so it can't improve the bit error rate characteristics sufficiently.In this paper, we propose the new ICI canceller based on MMSE which does notcause enhancement of noise component.Computer simulation confirmed that the ICI canceller based on MMSE canreduce the bit error rate under the Doppler-spread channel.
Studies on an application of digital terrestrial broadcasting to Intelligent Transport Systems (ITS) are continued. We conducted a field experiment to investigate the possibility of single frequency network (SFN) along the road using TAO ITS experimental facilities. Mobile reception characteristics were measured in a trial route and field strength distribution between transmitting stations was examined. The results show that SFN can be constructed by increasing the number of transmitting stations.
We developed a software defined receiver for digital terrestrial broadcasting, to apply it to multi-mode receivers used in automobiles. The receiver can define and update its function by downloading demodulation software. We describe the structure and function of the developed receiver. We show an example of downloading demodulation software that can improve receiver characteristics against multipath interference.
In terrestrial digital broadcasting, integrated internet and broadcasting services can be provided for the mobile user terminal including a cellular phone. In order for the experimental evaluation of the new service, a test bed terminal has been developed. This paper describes the total system configuration of the test bed terminal and its target application images.
This paper proposes the architecture of terminal for terrestrial digital broadcasting by cellular phone. This test bed terminal has been developed for the experimental evaluation of the fusion services of internet and broadcasting. This paper describes each function and management architecture of this terminal (hardware and software).
In digital terrestrial broadcasting service, internet and broadcast service for a mobile phone is considered. Then, a test bed terminal was developed for the purpose of experimental evaluation of this new service. Especially, this paper reports the broadcast reception function of the terminal.
Network-linked data broadcasting services combine digital terrestrial broadcast services targeting mobile terminals and Internet-based services. To test the effectiveness of such new linked services and prototype mobile terminals, we develop a terminal-function evaluation system.
Installing transmitting equipment for digital terrestrial broadcasting involves a great deal of expense in a very short period. Thus, a common goal of all broadcasters is to find a way to reduce this expense. To achieve this goal, NHK has been working together with all commercial broadcasting companies to find ways to reduce transmission equipment costs. These efforts have led to the publishing of a set of common-specifications for digital terrestrial broadcasting transmitting equipment. These specifications can be expected to lead to joint construction projects and reduced equipment costs through mass production, and further cost reductions achieved through joint maintenance management.
In order to build a digital terrestrial television broadcasting network, transmitter for main stations will be fixed in four years from 2003. Commercial broadcasting and NHK drew up the common-specifications(Orange Book), so as to hold down cost, securing reliability. This paper describes the outline of the digital transmitter specified to the Orange Book.
In digital TV broadcasting, easy maintenance and space saving transmitter is desired by transmitter operators. Also, high stability with output power and signal quality is required in a digital transmitter. To meet those requirements, the newly developed digital transmitter employs low profile PA design, central monitoring system and adaptive pre-correction technology. This paper shows the composition and the performance of this digital transmitter.
This papaer describes the outline of 10kW water-cooled transmitter system for ISDB-T. The output power of 10kW is obtained by combining two sets of 5kW transmitters.It has three sets of 5kW transmitters and one of three is used as a redundancy. Switching to the redandant transmitter is realized without any power reduction.
MCPA (multi channel power amplifier) is useful for Terrestrial Digital TV Broadcasting Transmitter. Several channel signals are amplified by one MCPA. This paper analyzes high efficiency and broadband performance of feed-forward power amplifier.
The detailed specifications for STL(Studio to Transmitter Link) / TTL(Transmitter to Transmitter Link) system for Terrestrial Digital Broadcasting Network have been discussed and developed in Working Group 2. Its transmission scheme consist two types, i.e. concerning TS and IF, this paper describes about the specification of STL/TTL system for terrestrial digital broadcasting network, especially about TS transmission scheme.
The overview of Standard Specification for primary station transmitting antenna, relay station transmitting antenna, relay station receiving antenna, and channel combiner to construct antenna systems in low cost for terrestrial digital broadcasting.
At a preparation of the transmission equipment for digital terrestrial television, there is a requirement to try moving an equipment investment back by means of a broadband antenna simultaneously to be able to transmit the channel that frequency departed . So far, though twin loop antenna had been much used as a transmission antenna for UHF, it was a narrow band and many kinds. We have developed a broadband 4L-type twin loop antenna(24U) .
For stable service of digital terrestrial broadcasting, the link budget of a relay network is very important. In this paper, one consideration is carried out based on the ARIB-STD and the equipment common specifications for digital terrestrial broadcasting about the link budget in consideration of various C/N degradation produced on a relay network.
We examined dependence of magunetic cluster size on read/write charactaristics and thurmal decay charactaristics in TbFeCo/CoCrPtB composive perpendicular magnetic recording media.It was found that composive media with the magnetic cluster size of 82nm shows both high media SNR and low thurmal decay.
A cylindrical magnetic storage system aiming at realization of high transfer-rate and high track-density was built as a trial. The system has the possibility of simplifying multiple-head control. Single-layered perpendicular medium was made on the cylindrical substrate and fundamental read/write characteristics using a commercial GMR head were obtained.
Medical images will be digitised electronically and shared over networks in the near future. This is expected to cause problems with storage-capacity and network performance. In this study, medical images were compressed and transferred over a network to investigate ways to reduce these problems. We found that compressing and sending still images individually was slower than combining a number of images into a single file and sending together. The improvement in efficiency resulted from reduced connection overheads.
We studied and designed a next generation multimedia content object model(WS-MCOM) using web service.WS-MCOM is a protocol for designing and developing a multimedia content on web service and WS-MCOM enables that a content works together other contents.In this paper, we show the outline of WS-MCOM.
Abstract As one of the challenges of closed captioning for live broadcast program, we succeeded in closed captioning of live baseball program using speech recognition. In this paper, we describe about the development of a speech recognition closed captioning system, and the results using the system.
The Telecommunications Advancement Organization (TAO) of Japan has initiated a project in order to develop the efficient production of closed caption in TV programs for the benefit of the hearing impaired people. In the phase-1 project, we have built an automatic closed caption production system and obtained plenty of suggestions for improving the system from its assessment. In the phase-2 project, we have been developing a new utility model of the system combined with manual procedures and automatic ones. We report the outline of the production system and the result of its practical assessment.
A speech recognition system with manual error correction has been developed in order to produce closed captions in live broadcasting programs such as news etc. In this paper, we propose an efficient correction system. In this system, just two operators can correct errors against two pairs in the current system, proposed system, however, is able to correct as keeping both high correction accuracy and short time delay as current system.
We developed a new real-time CG-animation system called "MIG (Motion Image Generator)" for TV casters and began to operate this system on TV programs (live sports/news).The use of "MIG" has made possible that TV casters control CG animation or change the view on CG-scene through the monitor with touch-sensor device. This paper describes the system and its effectiveness.
White Light Emitting Diodes, White LEDs developed by NICHIA Co. in 1993 have been used for various purposes such as large-scale color displays, instrument board in cars or back-light unit in small-sized LCD monitors. We have researched their possibilities as TV production light units while its emission efficiency is raised and come up to the mark. Redesigning a piece of LED device, we have achieved the high quality such as distribution of luminous intensity or color rendering property which is essential for TV production. We have developed Camera Light which is attached to camcorders and high power Video Light available with tripod stands.
The lowest practical luminous intensity under which conventional video camera can capture images is around 20-lx. The illumination available under a full moon in the night is 0.2-lx, which is approximately two orders of magnitude lower. Although a number of high-sensitivity imaging devices have been proposed including avalanche multiplication device, yet the sensitivity of these devices is still limited up to 30 times higher than that of conventional CCD. To achieve ultra high sensitivity to film astronomical phenomena, we have been working on an ultra-high-sensitivity HDTV color camera based on a 2.5-inch image intensifier since 1997. This camera can capture clear images of extremely dark objects in space, as well as of comets, nebula, and galaxies. However, its bulky size and heavy weight had been restricting its versatility. Ultra-high-sensitivity HDTV Color Camera has been developed featuring small-sized image intensifiers with GaAsP photocathodes, which provide very high quantum efficiency. To achieve HD performance and a compact camera body, we combined three 1-inch image intensifiers with a 2/3-inch taking lens and three 2/3-inch CCDs by means of a new optical system capable of enlarging and reducing images. Our approach well suited for compact implementations is to use a fiber optic plate instead of a relay lens. To use a 2/3-inch CCD, a FOP tapered from 1 inch to 2/3 inch was used to couple the image intensifier tube to the CCD. Because inconsistencies more easily occur due to graphic distortion and other causes with a tapered FOP than with a straight FOP, we had to address the prospect of registration errors in designing our camera. We solved this problem by selecting tapered FOPs from the same production lot and inspecting the best combinations by means of graphic comparison with a computer, so that their characteristics extremely matched. There is a critical requirement of incorporating a videocassette recorder system into a camera for operators to keep its simple and convenient mobility. It is realized due to the compact size, lightweight and little power consumption of these imaging devices. As a whole this camera weights 6.3-kg and consumes 48-W electric power including a VCR system. The camera provides excellent color reproducibility even under low light level conditions of 0.2-lx at an iris setting of f/2, with a signal-to-noise ratio of 55 dB at pedestal level. Its sensitivity is about 400 times greater than that of current HDTV CCD cameras, making it particularly well suited for capturing real time images of faint objects in space, aurora, etc., filming the nocturnal activities of animals in their natural settings, and reporting breaking news at night.
Rotating picture effects are often used for dramas, music programs. But to get rotating effects, it is necessary to use large equipment like crane camera. So we developed a camera which has rotating CCD block function. With this camera, directors and camera operators can get rotating picture effects easily.
It experimented by building the total system for performing The transmission of type _II_contents of telecommunication council report concerning the server type broadcasting and the technical verification of the access control technically. It was proven that transmitting large-capacity contents became possible by using the module link method and efficient transmission by VBR. It verified that the requirements for the server type contents of type _II_ like the access control by ACI were satisfying in the range of the telecommunication council report in general.
NHK opened the "NHK Archives" on February 1, 2003. The system is located in Kawaguchi City, Saitama Prefecture. It has two main components, the general database that manages the vast amount of stored video software, and the video library where the general public can watch video programs on demand.