Face recognition in 2D and 2.5D using ridgelets and photometric stereo

A new technique for face recognition – Ridgefaces – is presented. The method combines the well-known Fisherface method with the ridgelet transform and high-speed Photometric Stereo (PS). The paper first derives ridgelet projections for 2D/2.5D face images before the Fisherface approach is used to reduce the dimensionality and increase the spread of the resulting feature vectors. The ridgelet transform is attractive because it is efficient at extracting highly discriminating low-frequency directional features. Best recognition is obtained when Ridgefaces is performed on surface normals acquired from PS, although good results are also found using standard 2D images and PS-derived albedo maps.

► Face recognition using the ridgelet transform and photometric stereo.
► The method allows good dimensionality reduction while preserving edge singularities.
► Tests were performed on the PhotoFace database of natural poses and expression.
► Combination of ridgelet and Fisherfaces gave best results, especially on surface normals.