-
Hiroshi SANKOH, Sei NAITO
Article type: Article
Session ID: 6-5
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose a robust ball tracking method for multi-view videos shot in an outdoor soccer stadium. Our proposed method checks whether tracking is successful or not in each camera for every frame, and modifies ball region based on epipolar geometry between multi-cameras. Experimental results show the effectiveness of our method compared to conventional method using single camera.
View full abstract
-
Akio ISHIKAWA, Hiroshi SANKOH, Sei NAITO
Article type: Article
Session ID: 6-6
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, improvement scheme of 3D position estimation especially for occluded contours is proposed in order to achieve a precise shape reconstruction. To seek the high precision, the proposed scheme can utilize the sufficient surface points for the shape reconstruction by analyzing the tangent line from a camera principal point to the real surface.
View full abstract
-
Rie SAWAI, Kyoko ARIYASU, Hiroshi FUJISAWA, Shunji SUNASAKI
Article type: Article
Session ID: 7-1
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a TV program recommendation method using both location-based SNS and mini-blog type SNS. By combining multiple SNS whose characteristics are different, our method enables to estimate users' preference more precisely, and provide high performance recommendation service.
View full abstract
-
Yuko YAMANOUCHI, Makoto OKUDA, Masaki TAKAHASHI, Simon CLIPPINGDALE, M ...
Article type: Article
Session ID: 7-2
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
TV user interface called UTAN is proposed that makes TV viewing easier and more convenient. UTAN provides viewers with information presentation and recommendations by inferring the current viewing status.
View full abstract
-
Akiko SEKI, Pao SRIPRASERTSUK, Mutsumi SUGANUMA, WSataru KAMEYAMA
Article type: Article
Session ID: 7-3
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose a method of evaluating the personalized value of contents for content recommendation systems. In the proposed system, the user's interest from user's vital reaction and viewing history is dynamically detected and analyzed to define the value of contents based on both user-metadata and content-metadata.
View full abstract
-
Mutsumi SUGANUMA, Eri KAWAMURA, Wataru KAMEYAMA
Article type: Article
Session ID: 7-4
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this study, we would report the result of preliminary study on measuring the attitude against video content by user's bio-signals. In present study, we have measured the eye information and the background brain activity of viewers during watching video content. We would discuss the possibility of user's bio-signals for user's attitude estimation.
View full abstract
-
Hiroki SHIMADA, Kazuo OHZEKI, Yutaka HIRAKAWA
Article type: Article
Session ID: 7-5
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
From the game image of tennis, a continuation rally is analyzed and the construction of a system which gives leading information to a player is aimed at.
View full abstract
-
Eigo MIYAZAKI, Kazuo OHZEKI, Yutaka MIRAHAWA
Article type: Article
Session ID: 7-6
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Recently, the development of multimedia technologies such as the development of miniaturized mass storage medium speed and high-definition video transmission.But while memories become large, the user can say to get that information has not been considered. So reduce the burden on users to generate a highlight.
View full abstract
-
Yuki KANAYA, Chihiro FURUKAWA, Hideji ENOKIZU
Article type: Article
Session ID: 7-7
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In the present study, we proposed the mechanism of automatic video editing system that uses shooting rules on the basis of the movie technique. As a result, it was confirmed that our system can provide the video similar to the movie played in the theater.
View full abstract
-
Takuya SAITOH, Shunsuke MATSUMOTO, Mitsugu MATSUMOTO, Kuniya SATO, Rih ...
Article type: Article
Session ID: 7-8
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes the lecture contents with index by Using Ontology and its System.
View full abstract
-
Keiko AOKI, Hiromi ISHIZAKI, Akio YONEYAMA, Chihiro ONO
Article type: Article
Session ID: 7-9
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed an On the Fly Music Synchronized Slide Show System which shows pictures related to the lyrics synchronized to the music being playbacked. In this paper, we show how we developed the system as a realtime processing system.
View full abstract
-
Daisuke MATSUMOTO, Terumasa AOKI
Article type: Article
Session ID: 7-10
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose an effective parameters set using some image features and a filtering system which makes use of these parameters. We show that the proposed filtering system is a very powerful for the detection of adult animation, by experimental results.
View full abstract
-
Hiroki MORIGUCHI, Yoshihiro KURODA, Hidekazu TAKIUCHI, Shingo YAMAMOTO ...
Article type: Article
Session ID: 8-1
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a method for controlling the view of a camera using the flexible lens. The flexible lens was deformed by shape memory alloys and the view direction of a camera was changed.
View full abstract
-
Tatsuya Kobayashi, Haruhisa Kato, Akio Yoneyama
Article type: Article
Session ID: 8-2
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Conventional back-projection methods for camera tracking have problem when target objects have curved surface shape. This paper presents a novel back-projection method that can estimate precise 3d-coordinates and remove unreliable correspondences efficiently. Experimental results show that proposed method improves the performance of camera tracking for sphere objects.
View full abstract
-
Keita Okada, Takafumi Taketomi, Tomokazu Sato, Naokazu Yokoya
Article type: Article
Session ID: 8-3
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a robust camera parameter estimation method based on online feature landmark database (LDB) construction. In our method, LDB is automatically updated using panoramic images captured by a pair of omnidirectional cameras in the target environment. By using periodically updated LDB, we can cope with the visual changes in the environment for augmented reality.
View full abstract
-
Yang SONG, XiaoLin ZHANG
Article type: Article
Session ID: 8-4
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Kenta TOBISAWA, Terumasa AOKI
Article type: Article
Session ID: 8-5
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We are usually watching a movie while predicting the next scene. In this paper, we propose a new gaze model based on human expectation, which is extended by increasing the number of state and sticking labels on events.
View full abstract
-
Daishi WATABE, Hideyasu SAI, Katsuhiro SAKAI, Osamu NAKAMURA
Article type: Article
Session ID: 8-6
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
To deal with pose variation when ears are rotated in depth, Gabor jets of different poses are estimated and used as training data for a discriminant-analysis-based classifier. Experimental evaluations show the effectiveness of the proposed algorithm and the possibility of improving the robustness of a single-view-based ear surveillance-system.
View full abstract
-
Nobuyuki UMEZU, Shuichi HASEGAWA
Article type: Article
Session ID: 8-7
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Vectorization process of hand-drawn measured drawings of archeological remains is one of the most time-consuming tasks for researchers. A vectorization algorithm is proposed for converting scanned raster images into vector drawings suitable for publishing. The sectional patterns on a graph paper is eliminated using Fourier Transform and vectors are extracted by connecting local extrema of pixel values.
View full abstract
-
Tatsuya Ono, Nobuyuki Umezu, Masatomo Inui, Isao Shimizu
Article type: Article
Session ID: 8-8
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In the defect inspection of the part by using image processing, gaps are included between compared images. Recent digital cameras have much higher resolution, which makes the position gaps in inspected images greater. Even the situation including rotation and shift, this algorithm based on Fourier transform enables fast correction and defect detection.
View full abstract
-
Ryo NAKAYAMA, Yoshio YANAGIHARA
Article type: Article
Session ID: 8-9
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In recent years, image segmentation on scenery photos or nature photos has been one of the most significant issues. In general, for image segmentation, users are required to input some information presenting the difference between the regions of "object" and "background" in advance. In this study, the authors studied the automatic method for extraction of petal region on flower images by using hue component which no data input required.
View full abstract
-
Kenji HONDA, Makoto SATO
Article type: Article
Session ID: 8-10
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Previously, the zero-cross detection algorithm has been proposed. In this article, we define the lines of slope which are drawn from maxima to minima and propose the zero-cross detection algorithm for gray image by determining the concavity and convexity along the slope.
View full abstract
-
Takaki TSUCHIYA, Ichiro MATSUDA, Hisashi AOMORI, Susumu ITOH
Article type: Article
Session ID: 8-11
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes a video mosaicing method which generates a panoramic still image from video sequences shot by a handheld camera. The method automatically selects key frames and stitches them together. In order to reduce unwanted artifacts due to the stitching, the graph-cut algorithm is employed for seeking invisible seams in overlapped areas of the key frames.
View full abstract
-
Asahi SUZUKI, Yoshitsugu MANABE, Noriko YATA, Yuki URANISHI
Article type: Article
Session ID: 8-12
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes a new design of overlayable transmissive color marker. Each marker is transparent and the marker has different colors, and outside of the marker is a region for estimating order of the overlay. Moreover this marker can estimate the relative degree of rotation angle between markers.
View full abstract
-
Masahide KOBAYASHI, Yoshitsugu MANABE, Noriko YATA, Yuki URANISHI
Article type: Article
Session ID: 8-13
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In Mixed Reality(MR) it is very important to keep geometry, light, and time consistency. This paper proposes a method of real-time representing inter-reflection and shading effects by means of estimating the lighting condition using spherical mirror in MR using graphics processing unit(GPU).
View full abstract
-
Ryohei TANAKA, Naoki HASHIMOTO
Article type: Article
Session ID: 8-14
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose to use force and thermal sensation with wind for displaying virtual object's presence. Our proposal method uses a peltier device to control temperature on wind generated by a fan with a motor. When we touch virtual objects, we feel clearly their presence by the sensations from wind.
View full abstract
-
Masahiko Fujiwara, Terumasa Aoki
Article type: Article
Session ID: 8-15
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we will propose a new hypothesis that human beings possess a "reflection characteristics database for various kinds of foods" to perceive foods. Moreover, using perception characteristics of foods, we will make a study on quality improvement of food images.
View full abstract
-
Azusa GONOI, Tomoaki MORIYA, Tokiichiro TAKAHASHI
Article type: Article
Session ID: 9-1
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose an outdoor AR service technology that employs company logos painted on the wall and curing of buildings under construction as logos for AR application. This technology enables mobile AR without spoiling the scenery. It can distribute and advertise information of a building under construction such as finished appearance and tenants to be in the building.
View full abstract
-
Kohei YAMADA, Kazuo Ohzeki, Yutaka HIRAKAWA
Article type: Article
Session ID: 9-2
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
By image processing, shortening of a user's time is aimed at by extracting the highlight scene of Rugby automatically.
View full abstract
-
Kensuke IKEYA, Kensuke HISATOMI, Miwa KATAYAMA, Yuichi IWADATE
Article type: Article
Session ID: 9-3
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose the multiple robot camera system to capture moving subject in the large space, e. g. soccer player on the field, by panning the cameras cooperatively. We experimentally developed 3D models from their multi view images.
View full abstract
-
Yuki OKURA
Article type: Article
Session ID: 9-4
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Exhibit a variety of content in stereoscopic 3D TV and other equipment, a questionnaire asked whether the interest in content or equipment which subjects us. Analyze that what you are looking for a three-dimensional image by examining the results public.
View full abstract
-
Sei GEN, Hideki KAKEYA
Article type: Article
Session ID: 9-5
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes a 3D visual interface for vehicle driving based on coarse integral imaging. The proposed display can show parallax both in vertical and horizontal directions.
View full abstract
-
Yukio UEDA, Hideki KAKEYA
Article type: Article
Session ID: 10-1
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper we propose an integral volumetric imaging system with smooth motion parallax and moire-free volumetric expression by adding a fine fly-eye lens whose elemental lens has a long focal distance.
View full abstract
-
Shimpei SAWADA, Hideki KAKEYA
Article type: Article
Session ID: 10-2
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Coarse Integral Volumetric Imaging(CIVI) realizes multiview and volumetric display. One of the problems of CIVI is the trade-off relationship between the resolution and the number of views. We realize both using the lens array consists of decentered lens.
View full abstract
-
Yuya OKADA, Takanori SONODA, Hirotsugu YAMAMOTO, Shiro SUYAMA
Article type: Article
Session ID: 10-3
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose contrast improving method in 3-D image using multi-focus lens. The image quality of this 3-D display was degraded by ghost image which were caused by low contrast ratio of polarization-switching device. This degradation can be successfully improved by phase control of input and output lights using half-wave plate.
View full abstract
-
Shin MOTOKI, Shuichiro TAKIBANA, Hirotsugu YAMAMOTO, Shiro SUYAMA
Article type: Article
Session ID: 10-4
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We have investigated accommodation responses for a stereoscopic LED display by use of a parallax barrier. The results show that confliction between vergence and accommodation was partially reduced at 3.5-m viewing distance, and that accommodation to targets outside the depth of focus can be induced by vergence.
View full abstract
-
Naoki YAMADA, Hirotugu YAMAMOTO, Siro SUYAMA
Article type: Article
Session ID: 10-5
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The perceived depth in arc 3D display is estimated by changing the radius of arc and the illuminated angle. Perceived depth is in agreement with the diameter of arc when the illumination angle is 45 degrees. The slope of perceived depth dependence to the arc radius increases as the illumination angle decreases.
View full abstract
-
Nao SHIBUHISA, Kei TOKUI, Ken-ichi IWAUCHI, Sakae SAITO
Article type: Article
Session ID: 10-6
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose an adaptive disparity remapping method for reducing the cardboard effect. A simulation shows the result of having applied the proposed method to the 3D image taken by smart phone equipped with the twin camera.
View full abstract
-
Keiichi MORI, Hidezi ENOKIDU
Article type: Article
Session ID: 11-1
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We aims at showing clearly by what king of element shot length is decided in a movie. Shot size chosen as an element which determines a shot length. Although it was a part, the tendency was able to be discovered between shot length and shot size.
View full abstract
-
Tomohiro HASEGAWA, Hideji ENOKIDU
Article type: Article
Session ID: 11-2
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We assumed that shooting techniques help people to comprehend a film. Then we analyzed a sequence of shots using data mining technology. As a result, we constructed decision trees focused on the shot scale for automatic video editing.
View full abstract
-
Taichi SHIMAZAKi, Takehiro TAKEHIRO, Hidezi ENOKIZU
Article type: Article
Session ID: 11-3
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In the present study, we tried to construct a system that can generate animation from the story text automatically. Especially, in order to generate animation similar to animated film played at the theater, our system used shooting rules based on the shot analysis of several movies.
View full abstract
-
Daisuke KAJIYAMA, Hideji ENOKIDU
Article type: Article
Session ID: 11-4
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In order to comprehend the story text, readers construct the situation model that is assumed to be the dynamic image simulated in the mind. We propose a method that select some gist scenes of the story text and attach them as moving illustrations in it automatically.
View full abstract
-
Satoshi Ueno, Emi Myodo, Shigeyuki Sakazawa, Shirotomo Aizawa
Article type: Article
Session ID: 11-5
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a determination method of transition effects for image slide-show based on detected solid surfaces in the image. Experimental results show that it is effective especially in the case that images contain vanishing points or have eye gaze to solid surfaces.
View full abstract
-
Kohei Inoue, Kenji Hara, Kiichi Urahama
Article type: Article
Session ID: 11-6
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a non-iterative approximate solution method for two-dimensional principal component analysis (2DPCA). Experimental results on the ORL face database show the effectiveness of the proposed method compared with the conventional methods.
View full abstract
-
Simon Clippingdale, Mahito Fujii
Article type: Article
Session ID: 11-7
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We describe a set of novel person-independent deformable templates developed for face tracking applications.
View full abstract
-
Yuya SHIRAISHI, Kazumasa KOBAYASHI
Article type: Article
Session ID: 11-8
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We propose a network user interface that is suitable to deal with specific problems in a virtual network. These problems are different from the ones commonly found in a conventional network.
View full abstract
-
Akira EGAWA, Susumu SHIRAYAMA
Article type: Article
Session ID: 11-9
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We have proposed a novel image ranking method using eye tracking data. In this paper, we evaluate the effect of the visual saliency by comparing real eye movements and the synthetic eye movements based on the saliency map model.
View full abstract
-
Masaki TAKAHAHI, Masahide NAEMURA, Mahito FUJII, Shin'ichi SATOH
Article type: Article
Session ID: 11-10
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We have developed a method that estimates the relation between a TV viewer's attention and the viewed content. Body motion, time interval of eye blinking, and variations of eye gaze are used as features for the mental focus estimation. These features are extracted using an ordinary camera and Microsoft Kinect without any contact-type sensors.
View full abstract
-
Setiawan BONDAN, Takashi MATSUBARA, Kazumi MATSUMOTO, Tatsuya TOKUNAGA ...
Article type: Article
Session ID: 11-11
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper describes 3D gesture User Interface for table interactive display. The 3-dimensional gesture recognition algorithm and GUI design were optimized to create intuitive operation on the table-type display prototype.
View full abstract
-
Takahiko FUKINUKI
Article type: Article
Session ID: 12-1
Published: December 21, 2011
Released on J-STAGE: June 05, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS