All issues
Predecessor
33.55
Displaying 1-21 of 21 articles from this issue
- |<
- <
- 1
- >
- >|
-
Article type: Cover
Pages Cover1-
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (14K) -
Article type: Index
Pages Toc1-
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (83K) -
Masatoshi YAMAMOTO, Koji KIKUSHIMAArticle type: Article
Session ID: BCT2009-135
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIPTV service is widespread recently. First, download time and download quantity in a given time was measured when video file was downloaded simultaneously by PCs. The results show that download time with single PC is slower than multiple PCs when when the file formatl is downloaded with HTTP. Because single PC with HTTP downloaded at regular intervals. Switching time to another channel was measured. The results show that the switching time equalled buffer time plus one to three minutes when buffer time is ten secondes or more. When buffer time is ten or less, switching time is almost the same.View full abstractDownload PDF (525K) -
Yuichi KIDA, Takuro KIDAArticle type: Article
Session ID: BCT2009-136
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSFirstly, we present the optimum interpolation approximation for multidimensional vector signals. The presented approximation minimizes various worst-case measures of error of approximation simultaneously. Secondly, we consider a set of restricted multidimensional vector signals that all elements of the corresponding spectrum vector are separable-variable functions and we present the optimum interpolation approximation for the set of signals. Finally, by putting variables of the vector signals to be identical with each other, we present the optimum interpolation approximation of a generalized filter bank having generalized non-linear analysis filters. Finally, we present an application to environmental preserving problem.View full abstractDownload PDF (760K) -
Shouta SAKAINO, Hiromu KODAArticle type: Article
Session ID: BCT2009-137
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this paper, we propose a novel watermarking scheme for three-dimensional computer graphics (3-D CG) exploiting cross-correlation function in polar coordinate domain. First, we describe polar coordinate transform and its inverse transform. Next, we explain basic procedures for embedding watermarks into a 3-D model and detecting them from a watermarked model. Finally, we perform the computer simulation in order to prove the availability of proposed scheme. The experimental results for some test models show that the watermark information in each model can be detected by our scheme even when the watermarked models are subjected to attack such as random noise addition.View full abstractDownload PDF (788K) -
Shinichi SUMIYOSHI, Yuji SAKAMOTOArticle type: Article
Session ID: BCT2009-138
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSMethods which reconstruct 3D models from multi-view images come to the front due to recent advancement in image processing. There is a matter to be incapable of concavo-convex shape with spike in convention methods reconstructing 3D models from multi-view images. This paper proposes a method reconstructing 3D models from multi-view images of the object with concavo-convex shape and investigates a method to evaluate its models The proposed method calculates a field including proper surface of the object using Block Space Carving method and deforms the model in it. In deformation, the proposed method calculates a base surface of the object and deforms the model as a center of it. The proposed method reconstructs 3D models more accurate than conventional methods, by this processing. After deformation, the proposed method synthesizes 2D images at the camera positions the captured photographs and estimates accuracy of the 3D models. Replicating the deformation and the estimation accuracy of the 3D models, the proposed method reconstructs 3D models considering concavo-convex shape. At the end of this paper, the proposed method runs simulation experiments and experiments with photographed images. In the simulation experiments, the investigated method compares voxel data reconstructed from multi-view images with voxel data made from CG models data of the objects. In the experiments with photographed images, this paper evaluate the 3D models using PSNR. These results show that the proposed method is effective.View full abstractDownload PDF (1522K) -
Kota TAKEUCHI, Norishige FUKUSHIMA, Tomohiro YENDO, Mehrdad PANAHPOUR ...Article type: Article
Session ID: BCT2009-139
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe describe a method that generates free-viewpoint Images from recorded video by handy camera without any markers. A method that free-viewpoint image generation from a recorded video by handy camera was proposed in static scene. But the method needs a checkerboard design to calculate camera position/pose But in our approach, free-viewpoint images from recorded video by handy camera are generated using corresponding feature points. We calculate camera position/pose by factorization method. And we generate free-viewpoint images using camera position/pose by ray-space interpolation method.View full abstractDownload PDF (2264K) -
YOSHIHIRO SUGAHARA, MIKI YAJIMA, SHIN'ICHI TAKAGI, HIDEYOSHI TOMI ...Article type: Article
Session ID: BCT2009-140
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThe computer technology remarkablly advances now. A lot of products re invented along with it, and it is multifunctional and, in addition, has made it to high performance. Talking about television, digital broadcasting is spreading, and the data broadcasting is begun, too. On the other hand, it begins to be difficult for the users to operate them. this text describes the remote control system which is more accessable for users than now and the prototype system.View full abstractDownload PDF (1216K) -
MIKI YAJIMA, YOSIHIRO SUGAHARA, SHIN'ICHI TAKAGI, HIDEYOSI TOMINA ...Article type: Article
Session ID: BCT2009-141
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSUser-centric service is indispensable for the ubiquitus society. Because it is easy to acess contracted service from the arbitrary terminal or location. User-centric service is unrealizable in the present service contract system, unfortunately. The present service contract system, the only specific terminal that was granted ID can access the contracted service, has a low level of free selection for users. In this paper, we mention that it is necessary for the service in the ubiquitus society that has diversity of selection for the users and let the users use their services freely. We express it "user-centric service", and mention its requirements. And, for an example, we consider "user-centric content service".View full abstractDownload PDF (1228K) -
Masahiro Sekino, Terumasa Aoki, Junji NumazawaArticle type: Article
Session ID: BCT2009-142
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSMetadata for video content plays a very important role in many video related applications such as video retrieval, automatic video digest creation, copyright protection etc. However, metadata is not included in most of video content On the other hand, metadata can be automatically given for still pictures in Web pages (except for dynamic Web pages) by Web search engines. Therefore, if video content is associated with these still pictures automatically, it also makes it possible for video content to acquire its own metadata automatically. For this purpose, we propose a new matching method between video content and still pictures based on SIFT in this paper.View full abstractDownload PDF (1567K) -
Naoki Umeda, Terumasa Aoki, Junji NumazawaArticle type: Article
Session ID: BCT2009-143
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSOne of the fundamental image processing is shot boundary detection, but existing methods are not enough about their performances from practical points of view. We are researching a new shot boundary detection method based on frame-based clustering, which aims at decreasing false detection while keeping zero leakage (no overlooking of shot detection). In this paper, we propose a new clustering method in order to improve the performance, and report some experimental results about this method. In this method, each frame in video streams is regarded as one element in a multi-dimensional clustering space and this method detects shot boundaries accurately from this multi-dimensional clustering space by using a good clustering algorithm.View full abstractDownload PDF (1403K) -
Atsushi TANAKA, Xi ZHANGArticle type: Article
Session ID: BCT2009-144
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSMotion-JPEG 2000 has been adopted in Digital Cinema System Specification v.3 by DCI (Digital Cinema Initiative, LLC). Since Motion-JPEG 2000 does not use any inter-frame coding such as motion compensation, it is easy to edit and process movie. However, there is a drawback of flicker artifact in high-compressed movie. The flicker artifact is perceptible as small variation of brightness value due to lack of shift invariance of wavelet transforms. In this paper, we consider this flicker artifact as random noise, and propose. a new method for flicker reduction by using wavelet shrinkage in time direction. Also, we use a variety of videos to make a processing of flicker reduction, and evaluate the obtained experimental results subjectively and objectively. Furthermore, we discuss type of wavelet transforms, decomposition level and thresholding. Finally, we compare the results with the conventional method to show the effectiveness of the proposed method.View full abstractDownload PDF (1505K) -
Wataru OTAKE, Masahiro IWAHASHIArticle type: Article
Session ID: BCT2009-145
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn case of a two point rotation transform (RT), which is a component of the discrete cosine transform (DCT), is implemented in lifting structure, its multiplier coefficient has a singular point. Therefore, the coefficient value becomes close to infinity near the singular point depending on its rotation angle. As a result, rounding errors inside the transform are extremely amplified by such coefficients. In this report, we introduce permutation of order and sign of input and output signals of RT to shift the singular point. In this case, RT can be implemented with one of some candidates. Therefore, there exists the best combination of the candidates such that the error of the DCT becomes the minimum. We determine it utilizing structure of group of the permutations, comprehensively and efficiently. We confirmed that the error was reduced in Chen's DCT.View full abstractDownload PDF (601K) -
Ryosuke WATANABE, Koki ABEArticle type: Article
Session ID: BCT2009-146
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn H.264/AVC video coding standard, motion estimation is the most time consuming part to acheve accurate prediction and high compression efficiency. A fast motion estimation algorithm Unsymmetric-Cross-Multi-Hexagon-Grid Search (UMHEX) was proposed as a technique which saves computation time compared with Full Search. In this paper, we propose a method which further improves UMHEX with respect to the search speed without degrading image quality. Experiments on typical sequences defined under the common condition revealed that the search time of proposed method is 30% smaller than UMHEX in average with image quality almost unchanged.View full abstractDownload PDF (1022K) -
Kazuya HAYASE, Yukihiro BANDOH, Hiroshi FUJII, Seishi TAKAMURA, Kazuto ...Article type: Article
Session ID: BCT2009-147
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe have developed a full high definition video distribution system using H.264/SVC for multi-use video distribution services. This system consists of a software-based encoder, a stream extractor and a decoder which can process full high definition video data in real-time. Moreover, this system has a forward error correction function suitable for a H.264/SVC bit stream and an error concealment function based on a layered structure of the stream. This paper describes the whole structure of this system and the detail of the primary techniques, and shows the performance of this system.View full abstractDownload PDF (1122K) -
Satoshi OKADA, Tomohiro YENDO, Takaya YAMAZATO, Hiraku OKADA, TEHRANI ...Article type: Article
Session ID: BCT2009-148
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe consider an on-vehicle receiving system for long-distance and high-speed visible light communication. There are several problems associated with applying visible light communication to the field of ITS. It is necessary to receive information from long distance. And tracking the transmitter for a certain moving distance of the vehicle is also important. We applied an imaging optics to receive information over long distance, and two cameras are used to solve the relationship between the transmitter and the receiver position changes with time, and vibrational correction technique is also fixed to the system to minimize vibrational affectionsView full abstractDownload PDF (1273K) -
Daisuke UMEHARA, Masahiro MORIKURA, Toshiya HISADA, Shinichi ISHIKO, S ...Article type: Article
Session ID: BCT2009-149
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSControl signal networks in vehicles using dedicated cables such as twisted wires or fiber optics cables have been built for communicating between electronic control units (ECUs). However, there are problems such as the increase of the weight of the cable harness and the difficulty to insure the reliability of the complicated network architectures. Therefore, communication systems using in-vehicle power lines have been investigated. In our previous work, we proposed an impulse detection method by using two-state hidden Markovian-Gaussian noise model for noise on in-vehicle power line. However, the simulation results showed that the proposed method has low accuracy with the impulse detection when there are no impulses in observed noise sequences. In this manuscript, we propose information criteria to decide whether impulses are included in the noise sequences or not statistically and show that the proposed criteria improve the impulse detection accuracy by computer simulations.View full abstractDownload PDF (1112K) -
Takaaki Miyake, Hironori YamauchiArticle type: Article
Session ID: BCT2009-150
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this paper, we propose a method to model best suited to image's character and realize high-accuracy image decomposition by using Kalman Filter introduced from the model. Also, in realizing image decomposition, parameters of PSF are needed. Thus, we propose, in case of these parameters are unknown,, algorithms to estimate from an observed image including a decomposition process.View full abstractDownload PDF (1463K) -
Article type: Appendix
Pages App1-
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (86K) -
Article type: Appendix
Pages App2-
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (86K) -
Article type: Appendix
Pages App3-
Published: December 07, 2009
Released on J-STAGE: September 20, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (86K)
- |<
- <
- 1
- >
- >|