Block Selection in the Local Appearance-based Face Recognition Scheme
Hazım Kemal Ekenel, Rainer Stiefelhagen · 2006
In this paper we investigate the benefits of using a local appearance-based face recognition scheme against the problem of facial occlusion. We proposed two separate automatic block selection approaches to select the local image blocks that could be used for classification. Proposed approaches are tested against both upper and lower facial occlusions using the AR face database. Significant improvements are observed in the face recognition performance.