-
Article type: Cover
Pages
Cover1-
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Index
Pages
Toc1-
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Masahiro TADA, Haruo NOMA, Kazumi RENGE
Article type: Article
Session ID: ME2011-52/AIT2011-44
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In this paper, we propose (1) a method for continuously measuring and analyzing cyclists' turning-head behavior (including both safety visual search behavior and look-away behavior) using wearable sensors, (2) an automated method to identify whether the measured turning-head behavior is safety visual search behavior or look-away behavior requiring no prior information, and (3) a method for automatic creation of cyclists' safety behavior map showing traffic spots where many cyclists perform safety visual search behavior. Safety behavior map created from 36 junior high school students' data showed the fact that almost 60% of all subjects never perform visual search behavior even when entering blind intersections including accident prone area.
View full abstract
-
Yosuke YOSHIDA, Masahiro TADA, Haruo NOMA, Masaru NODA
Article type: Article
Session ID: ME2011-53/AIT2011-45
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The number of traffic accidents is on the decrease but it is still at high level. To prevent traffic accidents, we have proposed an automated method "HAS" for evaluating driving skill by measuring and analyzing safe driving behavior. HAS is calculated by using the sight distance depends on a geometrical feature of intersection. We confirmed effectiveness of HAS by comparing with evaluation given by driving instructor. Our experiment shows correlation coefficient between the evaluation results calculated by HAS and a driving instructor's evaluation is 0.71.
View full abstract
-
Akihiro OGINO
Article type: Article
Session ID: ME2011-54/AIT2011-46
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper proposes a model for user profile using Rough Set in order to develop a personalized information retrieval system. The model estimates attribute values of product that are related to user preference, using Rough Set. The model also calculates the degree of relation with the attribute values of product and the user preference. We estimate user preference with this method. The result shows that we can detect the attribute value related to user preference.
View full abstract
-
Taketo Kobayashi, Yusuke Iida, Tomoya Nii, Tomoya Ohneda, Akihiro Ogin ...
Article type: Article
Session ID: ME2011-55/AIT2011-47
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper presents a smart shop system that estimates preferences of consumers concerning products from their behaviors. Our method estimates the favorite clothes and their properties from direct and indirect interaction with the smart shop. Consumer's are grouped into 4 classes acording to their behavior patterns on the duration and frequency of "look", "touch", "take" and "watch" actions. Each consumer's preference on commodities is estimated acording to the average behavior pattern of his class. We examined our method through 20 persons ewperiments and showed 78〜87% accuracy on being interested.
View full abstract
-
Masashi MURAKAMI, Toshikazu KATO
Article type: Article
Session ID: ME2011-56/AIT2011-48
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This Research shows the modeling method of human auditory processes and its application. In modeling, authors show how human receive the multi-media contents, for example, music, and analyze the subjective evaluation scores and objective scores. Moreover, authors use this model for KANSEI services in the space that have multimedia contents, music, aroma, illuminations for users that have the image for the space.
View full abstract
-
Naoyuki Nakada, Yutaka Nakano, Toshiyuki Yoshida
Article type: Article
Session ID: ME2011-57/AIT2011-49
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
The authors have proposed a whole shape reconstruction technique, which integrates a set of 3-D models generated from each viewpoint by a Shape From Focus/Defocus (SFF/SFD) method. Since the SFF/SFD technique can estimate a measure of depth-estimation precision as a byproduct, the authors have also proposed a 3-D mesh integration technique using the measure as a weighting factor. However, the advantage of introducing the weighting factor has not been confirmed sufficiently because of a limitation of memory space. This paper thus aims at evaluating the impact of the weighting factor in the mesh integration, where errors between a computer-generated model and its reconstructed version are evaluated. The results show that the introduction of the weighting factor can reduce integration errors not only visually but numerically.
View full abstract
-
Daiki SHIRAISHI, Kazuhito MURAKAMI
Article type: Article
Session ID: ME2011-58/AIT2011-50
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper purposes a method to make a clip video automatically by using image features of the objects in the video. First, the objects in the video are extracted and that features "area", "number", and "position relations" are calculated. Then, image frames whose values of the features are more than the thresholds are selected. Experimental result compared with a clip video by manual shows that the performance of the purposed method is about 60%.
View full abstract
-
Yuta MISAWA, Kazuki YASUKAWA, Toshikazu KATO
Article type: Article
Session ID: ME2011-59/AIT2011-51
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
In web design productions, a person completes a design that is influenced by sensibilities of many different types of people. Therefore, we think it is necessary to make the sensibility visible, and have done researches in order to model people's sensibilities to designs. This text explains our system established by using the most influential colors to images and describes the research method.
View full abstract
-
Tsutomu IMBE, Kazuki YASUKAWA, Toshikazu KATO
Article type: Article
Session ID: ME2011-60/AIT2011-52
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
Color arrangement is considered to be a determined factor in graphic design. Base Colors and Accent Colors in Direct Message determine their overall image. But we have not clear mathematical definition of Accent Colors. This paper proposed mathematical definition of Accent Colors to given color arrangements. Based on this model, we can estimated the strength of the accent of the focal color. We adopt our model for printed materials such as Direct Message, and evaluated the efficacy of the estimation method.
View full abstract
-
Toshiaki TAKEDA, Hideaki KIMATA, Norihiko MATSUURA
Article type: Article
Session ID: ME2011-61/AIT2011-53
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
We have been studying the interactive panoramic video delivery method to aim at providing the highly realistic feeling for the user who is in the home as if he/she enjoyed watching in the event site, in which the whole event scene is distributed by a high resolution panorama video and the home user selects the viewing point and watches. In the paper, to achieve the more realistic feeling, we propose a novel multi-display viewing style and system based on the interactive panoramic video delivery method. Evaluation results on the efficiency of user's controlling the viewing point such as moving and zooming with multiple displays are also reported. Especially characteristics of the proposed style in terms of fast search of the subject at the start of viewing and solid pursuit during viewing are reported.
View full abstract
-
Takeshi OKUNAKA, Yoshinobu TONOMURA
Article type: Article
Session ID: ME2011-62/AIT2011-54
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This report introduces an interactive media system called Eyeke (Eye Mike)which converts video images into sounds based on feature level information included in the images. Eyeke is one of the steps toward creating interactive systems which make effective use of media's features. Indexed colors generated from hue values of video images are mainly used in the conversion. Eyeke works robustly by fine-tuned color indexing and calibration. Moreover, Eyeke can be used as a unique musical instrument with a hand-held camera which is like a baton used by a conductor. A user can 'play' music by interactively shooting musical notes printed on paper just in colors.
View full abstract
-
Takahiro ODERA, Masashi MURAKAMI, Toshikazu KATO
Article type: Article
Session ID: ME2011-63/AIT2011-55
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper discusses the relationship between image words and music feature quantity of popular music (Rock Music) songs. In our experiment, we selected eight image words, frequently observed in the user's evaluation in free term descriptions for music image. We obtained those image words based on Fourier analysis on music signals. We have applied our method to music retrieval from subjective image words, and shown the efficiency in the application.
View full abstract
-
Jun Ishibashi, Masashi Murakami, Toshikazu Kato
Article type: Article
Session ID: ME2011-64/AIT2011-56
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
This paper is a research about how the music and soundscape will effect to human's psychological stress. There're lot of sounds that it exists originallyin a space and it might be a stress to the person who is in the same space. It's difficult to adjust the sound itself but how about putting an another sound on it, think like there might be a masking effect and can decrease the psychological stress. This is an analysis of the music and feature value.
View full abstract
-
Article type: Appendix
Pages
App1-
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Appendix
Pages
App2-
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS
-
Article type: Appendix
Pages
App3-
Published: May 20, 2011
Released on J-STAGE: September 21, 2017
CONFERENCE PROCEEDINGS
FREE ACCESS