ITE Technical Report

[title in Japanese]

Article type: Cover
Pages Cover1-
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_Cover1

CONFERENCE PROCEEDINGS FREE ACCESS

Download PDF (15K)
[title in Japanese]

Article type: Index
Pages Toc1-
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_Toc1

CONFERENCE PROCEEDINGS FREE ACCESS

Download PDF (40K)
3D Shape Reconstruction from Multi-directional Microscope Images

Hiroki Ojima, Shogo Tokai, Yousuke Oota, Hiroyuki Hase

Article type: Article
Session ID: ME2009-86/AIT2009-71
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_1

CONFERENCE PROCEEDINGS FREE ACCESS

Show abstractHide abstract

In this report, we propose a method to reconstruct 3D shape of small object from microscope images. We use a microscope system which can observe the object obliquely and multi-directionally by a rotating mirror head. At first, a geometrical model of the microscope is derived. Next, we project each observed image to a plane assumed in a space using texture mapping. Then, each point on the plane can be evaluated about its validity by checking overlapping situation of projected appearance on the plane. By translating the plane parallel in the space, 3D shape is obtained as a set of valid points in thespace. In this report, color information and edge information is used as appearance features for this method.

View full abstract

Download PDF (1339K)
Subjective response to shape, lighting, and appearance for 3D furniture data

Hidetaka Takeda, Tatsuya Shibata

Article type: Article
Session ID: ME2009-87/AIT2009-72
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_5

CONFERENCE PROCEEDINGS FREE ACCESS

Show abstractHide abstract

We formulate a method for a new feature description of 3D object models. This method is based on combined descriptors with shape and specular reflexion to retrieve 3D objects with light. Light plays an important role for form and appearance of 3D objects. We did three-way ANOVA (ANalysis Of VAriance) for lighting, shape, and appearance. We obtain a good relationship between shape and appearance. We compare separate descriptors and combined descriptors for shape and specular reflexion by multiple regression analysis. As a result, we find that the position of specular reflection plays an important role in high-quality response to 3D objects with a spot light.

View full abstract

Download PDF (551K)
Extraction of focal points in visual searching

Tomoya OHNEDA, Toshinari KAMAKURA, Toshikazu KATO

Article type: Article
Session ID: ME2009-88/AIT2009-73
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_9

CONFERENCE PROCEEDINGS FREE ACCESS

Show abstractHide abstract

We propose systems for extracting fixations, microsaccades and focal points from eye mark data for visual searching. Through these observation, we can estimate the degree of interest to each object from the eye mark data.

View full abstract

Download PDF (895K)
An Optical Motion Capture System for Physical Motion Analysis in Driving

Shinji FUKUMA, Takeshi MAEDA, Ryosuke SATO, Masanori HANAI, Shin' ...

Article type: Article
Session ID: ME2009-89/AIT2009-74
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_13

CONFERENCE PROCEEDINGS FREE ACCESS

Show abstractHide abstract

This reports proposes an optical motion capture system for physical motion analysis in driving. Markers for motion capture consist of two high intensity LEDs, one for position measurement and the other for pilot signal. The pilot marker flashes synchronized with the periodic pulse, which is generated by multiplying the UTC second (1PPS) obtained by the GPS. Video cameras are also triggered with the pulse and thus the proposed system can distinguish the marker every one second in the automobile in the daytime. This copes with problems in optical motion capture system, occlusions and swappings.

View full abstract

Download PDF (957K)
Colorization technique for plain maps in the style of water color painting : A warping-based approach for existing colorized regions

Atsushi ITO, Toshiyuki YOSHIDA

Article type: Article
Session ID: ME2009-90/AIT2009-75
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_17

CONFERENCE PROCEEDINGS FREE ACCESS

Show abstractHide abstract

The authors have proposed a coloring technique for plain maps with a watercolor-painting style, which aims at a semi-automatic generation of, for example, city-planning images, conceptional drawings at completion, and so on. However, the technique still requires a certain amount of time and operations for coloring arbitrary-shaped areas such as mountains and rivers. This paper thus proposes a new coloring technique which utilizes a previously colored region as a texture and maps it to a target area by warping. The proposed technique can reduce the time and operations required for the coloring as well as increase a library of texture materials for map coloring.

View full abstract

Download PDF (1162K)
Automatic Image Classification Using Graphical Features and Keyword Evaluation with Tf-idf Method

Riki SHIGEMATSU, Takayuki UTSUGI, Toshikazu KATO

Article type: Article
Session ID: ME2009-91/AIT2009-76
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_21

CONFERENCE PROCEEDINGS FREE ACCESS

Show abstractHide abstract

Most of digital contents distributers use key words which correspond to objects in images to index various kinds of photo images. But these key words not always match with visual impression of images. We have developed a method to evaluate visual impression of images by using image key words which given by professional photographers. We have modeled their image evaluation process based on visual impressions (KANSEI Model). In this paper, we propose a method of automatic image classification using graphical features and keyword evaluation with Tf-idf method. By using the KANSEI model, we have developed automatic image classification system for various kinds of photo images based on visual impressions, and we have applied to similar image search system.

View full abstract

Download PDF (823K)
Collaborative Design Support System Using KANSEI Visualization : Application for Graphic Design

Takafumi INOUE, Yukihiko TAZAKI, Toshikazu KATO

Article type: Article
Session ID: ME2009-92/AIT2009-77
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_25

CONFERENCE PROCEEDINGS FREE ACCESS

Show abstractHide abstract

Graphic design such as Web pages is completed in KANSEI of many people crossing. Therefore, we think that the simulation of KANSEI is necessary, and have researched modeling KANSEI of a design. In this article, we report a collaborative design support system, and improve some problems of this system by various approaches.

View full abstract

Download PDF (903K)
[title in Japanese]

Article type: Appendix
Pages App1-
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_App1

CONFERENCE PROCEEDINGS FREE ACCESS

Download PDF (84K)
[title in Japanese]

Article type: Appendix
Pages App2-
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_App2

CONFERENCE PROCEEDINGS FREE ACCESS

Download PDF (84K)
[title in Japanese]

Article type: Appendix
Pages App3-
Published: May 25, 2009
Released on J-STAGE: September 20, 2017

DOIhttps://doi.org/10.11485/itetr.33.20.0_App3

CONFERENCE PROCEEDINGS FREE ACCESS

Download PDF (84K)

Register with J-STAGE for free!