Development of the dynamic system of analyzing and synthezising images of facial expressions in terms of image processing and computer graphics techniques, and its application for human behavioral researches.

利用图像处理和计算机图形技术开发动态分析和合成面部表情图像的系统，及其在人类行为研究中的应用。

基本信息

批准号：
01880007
负责人：
HARASHIMA Hiroshi
金额：
$ 7.42万
依托单位：
University of Tokyo
依托单位国家：
日本
项目类别：
Grant-in-Aid for Developmental Scientific Research
财政年份：
1989
资助国家：
日本
起止时间：
1989 至 1991
项目状态：
已结题

来源：
https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-01880007/
关键词：
Face Facial expressions Image processing Knowledge-based image coding Image synthesizing Computer graphics

项目摘要

The achievements of our research during last three years are listed as followings.1. Basic design of the system of analyzing and synthesizing inases of facial expressions.In the first place, the architecture of the system was examined. Secondly, softwares including ones for image processing and computer graphics to be necessary for developing the system were implenented into the graphic work station. Then, the method of systematic descriptions of inases of face, and facial expressions in terns of knowledge-based lmase codins approach developed by this head investigator and his co-workers. As a result, the advantage of the knowledge-based inase coding approach was verified.2. Development of application softwares for analyzing iiiages of facial expressions.The software of detecting the head notion and extracting parameters of shape concerning expressions was developed.3. Development of application softwares for synthesizing images of facial expressions.To synthesize natural images of facial expressions only in terms of parameters of shape concerning expressions, the way of deforming a model of face of the system was examined from psychological point of view, so that it can simulate real facial actions. Consequently, based on FACS, the generation rules for synthesizing images of facial expressions were constructed.4. Psychological AssessmentExtracted facial parameters of expressions and synthesized images of facial expressions were assessed by the psychological experiments. As a result, the validity of our developmental system of analyzing and synthesizins images of facial expressions were confirmed.5. Examination of the application of our' system for the researches in Psychology and Behay ioral Science.Significant researches on facial expressions and nonverbal communication in Psychology and Behavioral Science were surveyed. And the role of our developmental svstell to take in those research areas was examined.

在过去三年中，我们的研究成就被列为以下1。面部表情的分析和合成系统的基本设计。首先，检查了系统的体系结构。其次，将用于开发系统所必需的软件包括用于图像处理的软件和计算机图形，并将其隐藏到图形工作站中。然后，该主管研究员及其同事开发的基于知识的LMase Codins方法的面部界面的系统描述方法以及面部表情的方法。结果，验证了基于知识的INASE编码方法的优势。2。开发用于分析面部表情IIIAGE的应用软件的开发。开发了检测头部概念和提取形状的参数的软件。3。开发用于综合面部表情图像的应用软件。仅根据表达式的形状参数综合面部表达式的自然图像，从心理学的角度检查了系统面部模型的变形方式，以便可以模拟真实的面部动作。因此，基于FACS，构建了综合面部表情图像的生成规则。4。通过心理实验评估了心理评估的表达式和面部表情综合图像的面部参数。结果，我们的分析和合成素图像的发展系统的有效性得到了证实。5。研究了我们对心理学和行为科学研究研究的应用。对心理学和行为科学中面部表情和非语言交流的重要研究。研究了我们的发展SVSTELL在这些研究领域中的作用。