Face Recognition on 45-degree Pose Using Eigenface

Face Recognition have been the subject of interest for years. Various approaches were applied but most of the successful recognition systems are working with frontal views. The objective of the project is to examine the recognition performance of the chosen existing face recognition algorithm across 45-degree view pose such as how to handle the variability in face appearances due to changes in the viewing direction [1]. Using the eigenface method, the project is aiming to examine the performance of face recognition across 45-degree pose. The performance of chosen algorithm was analyzed in more detail. The algorithm then was tested against the photo database, observing result across 45-degree view pose. The failure and successes of the chosen approaches were analyzed and the results also have been discussed in this report.