Loading...
Please wait, while we are loading the content...
Similar Documents
The eigenface approach to automated face recognition.
| Content Provider | CiteSeerX |
|---|---|
| Researcher | Yee, Xiao Tan, Nikki |
| Abstract | Face recognition is a very high level computer vision task that can be achieved through various techniques. One of the most efficient and well-known techniques is the Eigenface approach. It involves obtaining a set of principal components, termed as eigenfaces, from a face database. These principle components are a compact representation of the face set. Any new face image can then be repre-sented as a linear combination of these eigenfaces. The work presented in this dissertation involved implementing the Eigenface algorithm for face recognition. The initial aim of this research project was to estimate the number of degrees of freedom in an eigenface recognition system through a series of experiments. Firstly, experiments testing the effects of image pre-processing and using different ranges of eigenfaces were conducted to inves-tigate the number of principal components needed for a good approximation of a face. Then, we conducted experiments to find the distribution of eigenface distances between different and same faces. We found that for different images |
| File Format | |
| Access Restriction | Open |
| Subject Keyword | Eigenface Approach Automated Face Recognition Principal Component Face Recognition Compact Representation Eigenface Algorithm Eigenface Recognition System Eigenface Distance Different Image Good Approximation Initial Aim Linear Combination Various Technique Well-known Technique Different Range Research Project Image Pre-processing New Face Image Principle Component High Level Computer Vision Task Face Database |
| Content Type | Text |
| Resource Type | Thesis |