2007 Fiscal Year Final Research Report Summary
Research on Facial Animation CG System beyond Three Dimensional Wireframe Model and Kansei Evaluation
Project/Area Number |
16300071
|
Research Category |
Grant-in-Aid for Scientific Research (B)
|
Allocation Type | Single-year Grants |
Section | 一般 |
Research Field |
Sensitivity informatics/Soft computing
|
Research Institution | Shibaura Institute of Technology |
Principal Investigator |
OHZEKI Kazuo Shibaura Institute of Technology, Faculty of Engineering, Professor (90317363)
|
Co-Investigator(Kenkyū-buntansha) |
OHKURA Michiko Shibaura Institute of Technology, Faculty of Engineering, Professor (00317364)
AOKI Yoshimitsu Shibaura Institute of Technology, Faculty of Engineering, Assistant Professor (00318792)
HARASHIMA Hiroshi The University of Tokyo, Graduate School of Interdisciplinary Information Studies, Professor (60011201)
HASHIMOTO Shuji Waseda University, School of Science and Engineering, Professor (60063806)
|
Project Period (FY) |
2004 – 2007
|
Keywords | 145-182facial image / wireframe model / feature point / synthesis / high-visopn / facial expression / eye-blinking / subjective evaluation |
Research Abstract |
In this project, results include increasing the number of feature points of wireframe model, extension of three dimensional wireframe model, improvements of motion data of facial expression, eye-blinking detection using a high-speed camera., several integrated texture images for a single facial part, slow-motion video of facial animation, kansei evaluation between synthesis of facial images and real facial images. For the eye-blinking detection using a high-speed camera, a motion pattern during eye-blinking interval is characteristic. Detection of the motion pattern is robust. For high definition, there are two items of special resolution and time resolution. For special resolution, a high-definition camera of 1280x720 progressive is introduced for eye region texture. For time resolution, a system of a high-speed camera is constructed using low-cost consumer video cameras to get 60-90 frames per second for eye-blinking motion. This process produces slow-motion playout for eye-blinking. Kansei evaluation is performed for two facial images of synthesized ones and really captured ones.
|