All issues
Predecessor
37.7
Displaying 1-33 of 33 articles from this issue
- |<
- <
- 1
- >
- >|
-
Article type: Cover
Pages Cover1-
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (10K) -
Article type: Index
Pages Toc1-
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (90K) -
Reiji WATANABE, Naoki HASHIMOTOArticle type: Article
Session ID: ME2013-1
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRecently, there are many reports of spatial display. We developed a balloon display with a fish-eye lens last year, but resolution, brightness and the focus of this system's image were insufficient. So, we propose a balloon display with multiple projectors for high-resolution and high-brightness image projection. And also, we propose a down-sizing implementation of this kind of multi-projection display system by using convex mirrors.View full abstractDownload PDF (7252K) -
Aya NAGAOKA, Naoki HASHIMOTOArticle type: Article
Session ID: ME2013-2
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRecently, the system using a projector and camera has appeared for some useful applications like geometric correction and 3D measurement. That system requires projector calibration for precise performance. However, current projector calibration methods take lots of time and effort, and some of their results are not so accurate. Therefore, in this research, we propose an easy and high-precision calibration using a depth sensor. First, the time and effort of the calibration is reduced by retrieving one-to-one correspondences between a projector image and a 3D space with the depth sensor. Furthermore, the accuracy of the calibration is improved by correcting the accuracy of depth value with the sensor.View full abstractDownload PDF (4543K) -
Satoshi OGAWA, Naoki HASHIMOTOArticle type: Article
Session ID: ME2013-3
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSA method of superimposed projection with multiple projectors which can project with higher resolution than using single projector has been focused. However, in previous method, all projectors are restricted to having same spec and being fixed at ideal position. Therefore, in this research, to reduce these restrictions, we propose a method of superimposed high-resolution projection that can use heterogeneous multiple projectors with arbitrary placement.View full abstractDownload PDF (10186K) -
Yuki Takai, Tomotaka NagaosaArticle type: Article
Session ID: ME2013-4
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this paper, we study on an indoor and outdoor AR campus guide system. This system consists of an outdoor guide system and an indoor guide system, and is implemented in the mobile terminal as a smart phone application. The outdoor guide system using the markerless AR can introduce our campus to visitors on the screen of mobile terminals. The information is about the names of buildings and the distance from the position of the building and the visitor's position. In order to achieve more detailed guidance, we add the system gallery to the guide system. We also create a wed page for new information of open campus. On the other hand, the indoor guide system uses the marker type AR. This system uses NyARtoolkit for Android, and implements two functions which are the playback mode and the video mode. We carried out the performance evaluation about the guide system using a questionnaire survey. The targets of the survey are students and open campus staff of our university.View full abstractDownload PDF (9501K) -
Masashi YUASA, Yasuyuki SAITOArticle type: Article
Session ID: ME2013-5
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThis paper shows an analysis of acoustic signals of music by natural observation method (NOM). NOM is a method of analysis of temporal waveform in real time and uses the current data and the past data. In this paper, we propose a beat tracking system which is based on NOM. First, transform acoustic signals to fundamental observation values (FOVs). Second, inverse-transform FOVs which contain only percussive component to a wave form. Then we obtained beat position.View full abstractDownload PDF (409K) -
Takahiro IIJIMA, Yasuyuki SAITOArticle type: Article
Session ID: ME2013-6
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThis paper shows a system which semi-automatically arranges an orchestra score for a duo score. To arrange an orchestra score for a duo score, we need to divide an orchestra score into solo and piano accompaniment score. However, it is too difficult to arrange an orchestra score for piano score. An orchestra score consists of several instrumental parts, therefore it needs to be arranged for piano accompaniment to play. This system arranges considered problems for playing piano and generates a piano accompaniment score. It combines a piano accompaniment score with solo part score manually extracted in advance, it generates a duo score.View full abstractDownload PDF (540K) -
Ryutaro NIHO, Yasuyuki SAITOArticle type: Article
Session ID: ME2013-7
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSBack Ground Music (BGM) is used to enhance the impression of the image. However, we have no method of automatic recommendation of a music that suits an impression of an image in currently. In this paper, we propose an automatic recommendation system of music matching input image. To recommend a music, the system extracts some features of both the input image and the music in database. And the system learns the correspondence of the image features to the music features using Canonical Correlation Analysis. We designed the system structure and verified the effectiveness of the proposed method.View full abstractDownload PDF (3537K) -
Yuri KANAMARU, Toshiyuki GOTOH, Yoshiro GOSHIMAArticle type: Article
Session ID: ME2013-8
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThis paper describes an algorithm for automated motion analysis of axonal transport, which is widely drawn attention to because the axonal transport is related to neurodegenerating disease and safety evaluation of drugs. However some methods were proposed, they have merits and drawback of sensitivity and accuracy on particle tracking. In this paper, we describe a method which consists of local tracking using kymograph and particle location images, and global tracking based on genetic algorithm, in order to solve these problems. The experimental results to evaluate effectiveness of our method are also presented.View full abstractDownload PDF (7958K) -
Yuya MUTO, Kazuhiro MATSUSHIMA, Mie SATOArticle type: Article
Session ID: ME2013-9
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSUsing indoor walls and floor as a projection surface, we can present surrounding images around a viewer by two projectors with fisheye lenses that are mounted on the ceiling of the room. In this paper, we perform the geometric correction to eliminate the distortion of the fisheye lenses from projected images. In addition, we fit the position of the ground in the projected images to the position of the floor in the room. As results, we can give a viewer a highly immersive feeling.View full abstractDownload PDF (13330K) -
Takuya TAKANOHAHSI, Yusuke AKIMOTO, Naoki HASHIMOTO, Mie SATOArticle type: Article
Session ID: ME2013-10
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRecently, since projection technology has been improved, image projection has become popular. For example, we pay attention to projection mapping which raises entertainment with projectors and interactive image projection in which viewers can participate. However, in most of the existing image projection, objects for projection are restricted to unmoving objects such as a screen and a wall. In this study, we propose image projection on moving objects by which viewers can freely interact with the projected moving objects.View full abstractDownload PDF (13006K) -
Shohei SAKAMAKI, Naoki HASHIMOTOArticle type: Article
Session ID: ME2013-11
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this research, we propose a new method of latency-free predictive projection for projection mapping on moving objects. Predictive projection uses images based on predicted movement of target objects, and is able to compensate for various latencies. In this method, we improve the prediction accuracy and reduce computational cost by focusing on visual critical areas. Moreover, generation of predictive images is implemented on GPU so that its update rate becomes to be 60fps, as extracting projector's maximum performance.View full abstractDownload PDF (5933K) -
Akiyoshi ITODA, Shin KOIDE, Shoji YAMAMOTOArticle type: Article
Session ID: ME2013-12
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSWe have developed an electronic note system to use digital pen and tablet PC. This system can make a digital note which can establish the link of web address and can embed the images from Internet. Moreover, it can perform group work by sharing the contents of digital note. However, if the previous digital content use in combination with actual paper sheet, it is difficult to handle information all at once. Therefore, we develop a presentation tool to superimpose between real paper and digital information to project back shot of digital contents by small laser projector. Since this system use CCD camera to detect the position of paper, the projection position is adjusted automatically according to the paper position. Moreover, this camera is available for image scanner when the real contents is necessary to use at the group work. In this presentation, we explain the basic function of our developed system, and evaluate the usability and benefit.View full abstractDownload PDF (14021K) -
Yukiko KAWAGUCHI, Naoki HASHIMOTOArticle type: Article
Session ID: ME2013-13
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRecently, virtual fitting systems with AR techniques have been used at some apparel shops. However they are very attractive and useful, it is not so easy and time-consuming to prepare clothes data corresponding to various postures in advance. Moreover, to create the 3D model which can reproduce realistic appearance and motion is also quite difficult. Thus, in this research, we propose a video-based virtual fitting system with which everyone can enjoy the fitting with various clothes and postures, by easily generating realistic appearances and natural motions of the clothes. Our system captures the clothes data from actual people with the clothes, and by combining that data with a skeleton model of target users, estimated with a depth sensor, we can generate various postured clothes data in real-time.View full abstractDownload PDF (10702K) -
Shuhei TOMITA, Keita TAKAHASHI, Masahide KANEKOArticle type: Article
Session ID: ME2013-14
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSMobile robots must be aware of the surrounding environment by the sensor. The robot cannot avoid a collision with a walker that appears suddenly from occluded area of vision sensor. In order to solve this problem, we calculate the occluded area and control the speed of mobile robot with the measurement results by LRF. The difference in vision depending on the movement path also affects the speed control. Quality of the routes, such as the shortest route, detour routes, and so on will be evaluated.View full abstractDownload PDF (13098K) -
Hitoshi MIYAZAKI, Keita TAKAHASHI, Masahide KANEKOArticle type: Article
Session ID: ME2013-15
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this paper, we propose a method for smooth human-robot interaction. The robot grabs human's hand when the robot guides for human. The robot can precisely convey its intention to human by taking a variety of actions in the grabbing hand. Also the robot can understand human's intention and request by recognizing the contact and the movement of human's hand. Our framework can achieve mutual communication between human and robot via haptics which is a key element of human robot interaction.View full abstractDownload PDF (8613K) -
Shin KOIDE, Akiyoshi ITODA, Shoji YAMAMOTOArticle type: Article
Session ID: ME2013-16
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRecently, real expression of material appearance is required in Computer Graphics. Especially, gloss reproduction is very important to reveal the property of material or/and smoothness of object's surface. The gloss appearance originated from specular reflection of light source. Sufficient amount of light is necessary to reproduce the real appearance. However, there is limitation in accentuation of gloss appearance only adding the amount of light. Therefore, we explore the possibilities of the accentuation by using movement of gloss according to the head tracking. In our system, observer's face is captured by CCD camera in front of screen, and the direction of the gaze point is estimated by the face aspect on the assumption that observer gaze the CG reproduction on screen. With this new function, we evaluate the reality of gloss appearance by varying the direction and gain of gloss movement.View full abstractDownload PDF (13688K) -
Koubun EI, Keita TAKAHASHI, Masahide KANEKOArticle type: Article
Session ID: ME2013-17
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRobots that explain something to people are actively researched for the use of museum guide, etc. However, the flexibility of interactions was limited because these robots were often implemented as non-human characters such as stuffed animals. Moreover, multi-user support and bidirectional communication with users were difficult with these robots. In this paper, we aimed to make a humanoid robot to conduct effective presentations by using its body gestures, hand gestures, and virtual laser pointers, as done by human presentators. Observing users through cameras, the robot should change its behaviors according to the states of the users, e.g. the robot should provide a more detailed explanation if people's understanding seems to be insufficient. This paper describes the development of our presentation robot system, and reports several evaluations conducted with small audience groups.View full abstractDownload PDF (9928K) -
Kazushi NAKAZAWA, Keita TAKAHASHI, Masahide KANEKOArticle type: Article
Session ID: ME2013-18
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSOur research is focused on mobile robots that can accompany a person, and this paper addresses how to control the relative position of the robot to the accompanied person according to the dynamic environment. The robot is expected to move side-by-side with the person in the normal situation, but the position in front or behind the person might be better if there are some obstacles. The shape of the artificial potential field of the accompanied person is devised to smoothly control the robot position in a unified way. Laplace potential method is also used supplementary to cope with extremely cluttered environments. Simulations and experiments in real space show the usefulness of the proposed method.View full abstractDownload PDF (17556K) -
Yusuke AKIMOTO, Mie SATOArticle type: Article
Session ID: ME2013-19
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSScenes where images are projected on non-special screens, then, the distortion by the shape of the projected surface occur to the images. It is necessary to make geometry correction on input images to remove the distortions. This study proposes a high-speed geometry correction technique using a depth sensor which spreads recently. By acquisition of the shape of the projected surface and making geometry correction on input images dynamically, images can be projected without the distortion for the viewer without depending on the shape of the projected surface. In addition, if during the projection, the positional information of the viewer is acquired, and geometry correction can reflect the positional information, the viewer can appreciate the projected images at a free position.View full abstractDownload PDF (10110K) -
Kazuhiro MATSUSHIMA, Yuya MUTO, Mie SATOArticle type: Article
Session ID: ME2013-20
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSImmersive image projection has been studied that gives an immersive feeling by projecting images around a viewer. Immersive projection displays are large-scale and expensive systems. In order to project images around viewer, this study projects immersive images by installing a fish-eye lens in front of a projector. However, it is difficult to remove the distortion by the fish-eye lens on the projected plane. The distortion by the fish-eye lens becomes stronger near the edge of the projected plane. In this paper, we perform various geometric correction methods and project immersive images without the distortion.View full abstractDownload PDF (13414K) -
Ayano HOSOGAI, Naoki HASHIMOTO, Mie SATOArticle type: Article
Session ID: ME2013-21
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSRecently, we have paid attention to projection mapping by which images are projected on walls of buildings since projectors have advanced in high brightness. In projection mapping, creators visually adjust geometric and color on projected images. The adjustments take hours, but adjusted color cannot correspond change of ambient lights. In this paper, we examine projection mapping with dynamic luminance correction that removes undesired influence by projection environment from projected images.View full abstractDownload PDF (9066K) -
Takeyuki ISHII, Hitomi MURAKAMI, Atsushi KOIKEArticle type: Article
Session ID: ME2013-22
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this paper, we propose a new method of feature homogenization to prevent changes in the features according to human subtle direction change, in the detection of the human behavior of CHLAC. The conventional human behavior detection using CHLAC, it shows a high performance, but the situation that moving greatly changes the direction of the vector, abnormal operation normal operation and different operation if it is difficult to determine. We propose a method in addition to improved processing mask pattern based on the correlation and vector averaging process based on an amplification to suppress the change in the amount of features according to the direction of movement of the person. This provides a robust human motion specific person to change direction.View full abstractDownload PDF (492K) -
Aishan TUERXUN, Zhiyu CHEN, Shogo MURAMATSUArticle type: Article
Session ID: ME2013-23
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn order to realize an efficient remote control of microscope for medicine, a lossless/lossy image sequence codec system is proposed by adopting motion-compensated temporal filtering (MCTF) and Motion JPEG2000. The proposed system exploits information of translation command as a motion vector given from the remote operator for searching a region of interest, and applies a global motion-compensation (GMC) to even frames. The GMC frames are used to predict odd frames in the temporal lifting structure and the difference is transmitted as an enhancement layer. In the static case, the base layer consisting of even frames and enhancement layer consisting of the difference are completely used to realize the lossless decode. On the other hand, when searching a region of interest, only the base layer is employed so that lossy decoding is realized for reducing the transmission cost. As a result, the usability is improved. In this report, the basic structure is investigated and verified the significance of the scalability function according to the static and dynamic operation case.View full abstractDownload PDF (4027K) -
Hai YAO, Norihisa TERASAWA, Shogo MURAMATSU, Takamasa SUZUKI, Toshiro ...Article type: Article
Session ID: ME2013-24
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSA metal sheet inspection system with Kinect is introduced in this work. The current sheet metal inspection system which the authors have proposed consists of an inspecting table and operation terminal. Since the table and terminal are separately installed, it is obliged to useless operations. In order to solve this problem, it is proposed to introduce a gesture recognition function of the Kinect sensor and improve the usability. A depth image is acquired by using Kinect, and then a fundamental gesture recognition is realized by using the OpenNI function. The fundamental gestures are considered to be adopted to the inspection operation terminal and the gesture operation on the inspecting table is discussed.View full abstractDownload PDF (4018K) -
Takahiro ISHIKAWA, Masashi OKUDAIRAArticle type: Article
Session ID: ME2013-25
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn this article, a computer based program chart developing method is proposed. Hierarchical compact description chart (HCP chart) has been proposed for upper process software development. Originally, it was drawn with hand. A script based drawing method has been proposed which translates directives into HCP chart graphics. In this article, GUI based drawing method is proposed based on HTML5 and JavaScript. HCP chart is constructed with symbols and relation lines. Symbols denote each role in a process, such as processing unit, iteration, switch condition, and input/output data elements. Their relations are shown with lines. All components and their relations of HCP chart are put on the graphic window directly. Thus, users design HCP chart easily and can intuitively understand total process. Data sharing functions are also implemented for collaborative software development.View full abstractDownload PDF (512K) -
Gibran Benitez-Garcia, Gabriel Sanchez-Perez, Hector Perez-Meana, Keit ...Article type: Article
Session ID: ME2013-26
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSThis paper presents a facial expression recognition algorithm based on segmentation of a face image into four facial regions (eye-eyebrow, forehead, nose and mouth) in order to develop a facial expression recognition under partial occlusion. As the special case of partial occlusion, half face occlusion is also examined. In feature extraction stage the proposed method employs Sub-block Eigenphases algorithm that uses Principal Component Analysis (PCA) and for classification stage it applies the Support Vector Machine (SVM).View full abstractDownload PDF (4966K) -
Kento Ishizuka, Ryosuke Nakamura, Toshiyuki Gotoh, Naoyoshi Tamura, Hi ...Article type: Article
Session ID: ME2013-27
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn orchestra performance, performers interactively exchange the musical information parameters, such as velocity and pitch under the conduct of a conductor. In this research, a multi-agent model of orchestra system is first proposed, in order to simulate interactions among performers. Then, we describe the developed prototype system that enables users to take part in as a conductor or a player. The extension of MIDI standard to transfer the articulations for the system is also discussed, in order to track score information correctly.View full abstractDownload PDF (9172K) -
Yuuki Iwasa, Daisuke TAKAHASHI, Noriyoshi OKAMOTOArticle type: Article
Session ID: ME2013-28
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSIn recently years, mobile phone or smartphone has been equipped with a biometrics of advanced security. In generally, iris certification systems use special device infrared camera. However, it has been equipped with infrared camera. This research aims at development iris certification systems for visible camera. In this paper, using processing of HOG (Histograms of Oriented Gradients), some types of features is extracted from input images by visible camera. The values of each the angle components obtained from HOG used to determination for each iris features. Therefore, the effective of the proposed technique was shown.View full abstractDownload PDF (9892K) -
Article type: Appendix
Pages App1-
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (79K) -
Article type: Appendix
Pages App2-
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (79K) -
Article type: Appendix
Pages App3-
Published: February 09, 2013
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS FREE ACCESSDownload PDF (79K)
- |<
- <
- 1
- >
- >|