-
Article type: Cover
Pages
Cover1-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Index
Pages
Toc1-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Bibliography
Pages
Misc1-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Tatsuya Ueda, Yasuhiro Takaki
Article type: Article
Session ID: 3DIT2011-51/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this study, the viewing-zone scanning type super multi-view display is developed. The conventional viewing-zone scanning type super multi-view display has a viewing region where an entire three-dimensional (3D) image cannot be observed. With the conventional system configuration, an image is projected with parallel rays, and is scanned horizontally by a galvano scanner. With the system configuration proposed in this study, an image is projected with rays converging at the distance where view points are generated. The viewing region where an entire 3D image is observed is maximized at the ray converging distance. In this study, the prototype system with 132 view points is demonstrated.
View full abstract
-
Takahiro KAMIKURA, Yasuhiro TAKAKI
Article type: Article
Session ID: 3DIT2011-52/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The two-view multi-view conversion technique generates a multi-view image from a two-view image. This technique is composed of the depth estimation and the parallax image generation. In this study, we explore the depth estimation method that is suitable to the real time two-view multi-view conversion. The simple stereo matching method, the coarse-to-fine (CTF) method, and the α-expansion method are compared; these three methods are representative depth estimation methods. We found that the CTF method requires the shortest calculation time and has robustness to a coarse texture region.
View full abstract
-
Naoki KAWAI, Akihiro SASAKI, Takahiro MATSUBARA
Article type: Article
Session ID: 3DIT2011-53/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
It is widely used in 2D/3D conversion to construct 3D meshes from an original image and a given depth map, and to render multi-view images from constructed meshes. Here, occlusion correction is practically essential that includes detecting occlusions and filling them with appropriate color with continuity in borders of the regions. In this report, we present a simple and practical method for occlusion correction by introducing label map, which records the identifiers of objects, in addition to depth map.
View full abstract
-
Kenji YAMAMOTO, Yasuyuki ICHIHASHI, Takanori SENOH, Ryutaro OI, Taiich ...
Article type: Article
Session ID: 3DIT2011-54/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We designed and manufactured 300-camera array. Each camera has 1600x1200 pixel and is located on the circle with 1.2 degree interval that is about 32mm interval. The direction and color balance of each camera can be adjusted by mechanical attachment and software. We generated hologram using the images captured by this camera array.
View full abstract
-
Hiroki BANDO, Shiro SUYAMA, Hirotsugu YAMAMOTO
Article type: Article
Session ID: 3DIT2011-55/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Our goal is to realize a floating digital signage by use of an LED panel. In order to form a floating image, a reflective imaging device, called RMGA (Roof Mirror Grid Array), is investigated. RMGA is suitable for large LED signage because of its scalability and robustness. Unlike RMGA for an LCD panel, RMGA for an LED panel is required to have a high aperture ratio in order to complement the black area of the LED panel. We have fabricated RMGA with an aperture ratio higher than 0.5. Experimental results show convergence of a light source to its mirror image position with a certain extended size.
View full abstract
-
Hiroshi SONOBE, Kei SADAKUNI, Hirotsugu YAMAMOTO, Shiro SUYAMA
Article type: Article
Session ID: 3DIT2011-56/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Protruding DFD (Depth-fused 3-D) perception has been evaluated by changing the distance between front and rear plane. Perceived protruding distance decreases as the distance between two planes increases. The distance protruding backward from rear plane can be perceived to be larger than the distance protruding from front plane.
View full abstract
-
Atsushi YAMANAKA, Natsuko IKEDA, Takashi KANEKO
Article type: Article
Session ID: 3DIT2011-57/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
To assess biological effects of 3D images, it is considered effective to analyze biological parameters such as heart rate, blood pressure, etc. In this paper, we evaluated the pulse wave of the viewers while looking at the test sequence including 2D and 3D images. The results indicate that there are significant differences in the variance of the indexes called rho-max between the condition of 2D images and 3D images.
View full abstract
-
Yuki ISSHIKI, Hirohiko KANEKO, Makoto INAGAMI
Article type: Article
Session ID: 3DIT2011-58/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We can perceive 3D structures in visual field from binocular disparity. However, the tolerance in the difference of the images for stable 3D perception is not clearly defined. For example, when there is a remarkable inconsistency in binocular images, the problems such as "eyestrain" and "inappropriate depth perception" could occur. In this study, we investigated the effect of the difference in the spatial frequency components of the binocular images on depth perception. The results indicate that high spatial-frequency components of the image are important for deciding stereo acuity.
View full abstract
-
Yoshitomo NAKAMURA, Masato NAGASAWA, Hironobu YASUI, Nobuhiko YAMAGISH ...
Article type: Article
Session ID: 3DIT2011-59/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We evaluated the binocular parallax of fusion and separation for 3D display. There is a difference in the fusion parallax angle and the separation parallax angle even if it watches the same stereoscopic picture. The case of the parallax angle in front of display surface is 4.35 degree in fusion and over 12 degree in separate from this evaluation result. The case of the parallax angle behind display surface is 2.55 degree in fusion and 3.51 degree in separate from this evaluation result. Moreover, the method of measuring the time to obtaining the binocular vision was examined. The tendency to become shorter was obtained at the fusion time when the stereo image is moved in the direction in the back.
View full abstract
-
Yuki MAEDA, Daisuke MIYAZAKI, Satoshi MAEKAWA
Article type: Article
Session ID: 3DIT2011-60/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We developed a volumetric display that uses a rotating prism sheet for scanning and a dihedral corner reflector for imaging. Since mechanical load to rotate a prism sheet is less than the load to drive a galvanometer scanner, a rotating prism sheet can scan a larger area than a galvanometer scanner. A dihedral corner reflector forms an image in the air. We developed a prototype volumetric display and displayed a volumetric image that has a size of approximately 5[cm]×5[cm]×5[cm].
View full abstract
-
Daisuke MIYAZAKI, Kenta OKODA, Takaaki MUKAI
Article type: Article
Session ID: 3DIT2011-61/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This report describes the proposal of a three-dimensional display system based on the combination of integral imaging, 360-Degree scanning with a rotating mirror, and imaging in the air with a concave mirror, aiming at floating three-dimensional image generation for full-parallax autostereoscopy, which can be observed from surrounding area. The proposed method is verified by preliminary experiments.
View full abstract
-
Mao MAKINO, Roberto Lopez-Gulliver, Shoichiro IWASAWA, Masahiro KAWAKI ...
Article type: Article
Session ID: 3DIT2011-62/IDY2011-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper aims to provide necessary guidelines for contents creation for auto-stereoscopic displays with dense horizontal parallax. First, we investigate the resolution characteristics of the displayed 3D image in auto-stereoscopic displays. Based on these resolution characteristics, we identify the region with best resolution to effectively display 3D contents. In the case of regions with low resolution, we suitably process the images in order to avoid the resolution loss due to the display's special configuration, thus achieving a more natural image viewing for the observers. Experiments results, using our in-house 200-inch auto-stereoscopic display, show natural 3D images especially suitable for its resolution characteristics.
View full abstract
-
Article type: Appendix
Pages
App1-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Appendix
Pages
App2-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Appendix
Pages
App3-
Published: June 09, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS