- 【Updated on May 12, 2025】 Integration of CiNii Dissertations and CiNii Books into CiNii Research
- Trial version of CiNii Research Knowledge Graph Search feature is available on CiNii Labs
- Suspension and deletion of data provided by Nikkei BP
- Regarding the recording of “Research Data” and “Evidence Data”
AN ILLUSTRATION DRAWING SYSTEM REFLECTING IMPRESSION OF A FACE
-
- KATAOKA Maki
- Keio University
-
- HAGIWARA Masafumi
- Now working for NTT Data corp
Search this article
Description
In this paper, we propose an illustration drawing system reflecting Kansei, the impression of the face. The system has the following advantages: a face in the illustration looks like the person in theoriginal picture; it can emphasize the person's impression; illustrations with expression can be obtained; whole body illustrations can be obtained; direction of the face can be adjusted in accordance with the body part; it has excellent interfaces; it can reflect user's sensitivity; and change of clothes is easy. In the system, first, a skin color region, a face part region and the edges are extracted from the inputted reference face image taken by a CCD camera. Position and shape of eyes, eyebrows, nose, mouth and jaw in the image are extracted. Next, a face illustration with emphasized characteristics is drawn. Face expression is given to the illustration by using a face parts database in the system. Each part has a value for facial expression calculated from many questionnaires. Then, the hair type can be changed to keep a harmony of the body and the face. The face part is connected to the body part illustration which a user chooses. In addition, a user can select glasses from a glass database and they can be added to the illustration. The effectiveness of the proposed system is confirmed through various evaluation experiments.
Journal
-
- KANSEI Engineering International
-
KANSEI Engineering International 3 (2), 7-16, 2002
Japan Society of Kansei Engineering
- Tweet
Details 詳細情報について
-
- CRID
- 1390282679461325312
-
- NII Article ID
- 130004301674
-
- ISSN
- 18845231
- 13451928
-
- Text Lang
- en
-
- Data Source
-
- JaLC
- Crossref
- CiNii Articles
- OpenAIRE
-
- Abstract License Flag
- Disallowed