1996 Fiscal Year Final Research Report Summary
Image Processing for detection and tracking of facial features, and User orientaed visual communication
| Project/Area Number |
07455155
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
情報通信工学
|
|---|
| Research Institution | Univ. of Tokyo |
|---|
Principal Investigator |
HATORI Mistutoshi Univ. of Tokyo, Dept. of Information and Communication Eng., Professor, 大学院・工学系研究科, 教授 (60010790)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
AIZAWA Kiyoharu Univ. of Tokyo, Dept. of Information and Communication Eng., Associate Professor, 大学院・工学系研究科, 助教授 (20192453)
|
|---|
| Project Period (FY) |
1995 – 1996
|
| Keywords | Image Processing / Eye contact / Detection of facial features / Tracking of facial features / アイコンタクト |
|---|
| Research Abstract |
In this research, various end user equipment and related technologies required in developing advances visual communication systems and thier impact on effective interpersonnel communication are investigated with some related image processig techniques such as face feature detection, feature tracking atc, . Those are needed to fully automate such systems. In particular following ideas are proposed.
MPEC (A novel teleconferencing system, capable of providing multiple person eye contact)
Here a novel method of providing eye-contact for multiple participnats in a teleconferencing system is presented. The method proposes a simple idea to full-fill the requirement of eye contact, for two or more local and remote participants simultaneously, by using half mirros and cameras placed at the commom points of the extended lines of gaze of each participant. We observed the resultant images by using a preliminary experiments and also building a prototype system. It gives any participant a feeling of "being looked at" if any remote participant are looking at him. On the other hand, he gets a feeling af "not being looked at" if non of them are not lookinf at.
Face Feature Detection and Tracking
Detection and tracking of facial features without any head mounted devices may become required in various future visual communication applications, such as teleconferencing and virtual reality ets, . We proposed automatic method of facial feature detection using a method called adge pixel counting. The proposed method utilized the edge information to estimate the facial feature positions such as eyes, nose, and mouth in the first frame of the sequence. Feature tracking is carried our alternatively using a method called deformable template matching and edge pixel counting.
|
