This is one of the core projects of Morishima-lab. in which Future Cast System is drastically updated and advanced to be a very affordable technology. A face geometry estimation method with light calculation cost and high accuracy only from a single picture using face prior from a collected face database is proposed at first. Then from video captured face image, a much more accurate 3D face geometry can be reconstructed by utilizing a Structure-from-Motion method to calculate 3D coordinates of sparse feature points as well as a former 2D feature points and face priors of frontal face image. And now, new trial is performed to achieve a higher depth accuracy by utilizing a normal information calculated by Shape-from-shading or by utilizing only a diagonal face silhouette curve information.
Face rotation angle adjustment method to achieve a higher performance of face authentication system and also an accurate skull fitting method to a head part of face image to estimate a stable local coordinate when facial expression change are composing a recent our hot topics. Based on these basic techniques, applications for a fat and slim simulation of face image by skull fitting method, an accurate face landmark point detection with small database and a highly accurate detection of an early stage of car driver’s drowsiness in a dashboard mounted video are also included in this project. Recently, a modeling of personal characteristics in face is focused and tackled strongly to improve a face aging performance and to control a real variation of attribute in face.
This project is supported by Ministry of Education, Culture, Sports, Science and Technology and OKI cooperation.
- 補助事業名：文部科学省 科学研究費補助金
A location of skull is estimated based on MRI database. The slim-fat simulation is realized by controlling a thickness of a fat layer.