2001 Fiscal Year Final Research Report Summary
Content-based Image Retrieval and Coordination Based on Subjective Criterion.
Project/Area Number |
10835018
|
Research Category |
Grant-in-Aid for Scientific Research (C)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
感性工学
|
Research Institution | CHUO UNIVERSITY |
Principal Investigator |
KATO Toshikazu Chuo University, Faculty of science and Engineering, Professor, 理工学部, 教授 (50297107)
|
Project Period (FY) |
1998 – 2001
|
Keywords | Kansei model / content-based image retrieval / similarity retrieval / statistical learning / multidimensional analysis method / VRML / color coordination / color perception |
Research Abstract |
(1) Computational modeling of subjective criterion in similarity on complex objects : We have investigated mutual interactions in perception process on various factors, such as colors, textures, shapes, spatial arrangements of 3D objects, which feature complex objects in human information processing mechanism. We developed integrated model on color and texture perception by generalizing our computational model on brightness perception. (2) Computational modeling of subjective criterion in similarity on complex scene : We applied the ideas above to the perception processes for scenic images, such as landscape images, interior images, which are composed of many objects. We also applied them to color coordination tasks, which are popular in industrial design, interior design, dressing. We developed similarity retrieval algorithms as well as coordination algorithms based on the simulation of color perception on the collection of coordination data. (3) Integrated user interface for content-based retrieval and coordination : We also developed integrated user interface for content-based retrieval and coordination of space design tasks, such as interior design, garden planning. The interface is built on VRML and web protocol which enables easy-to-use environment through the internet.
|
Research Products
(14 results)