2D automatic face landmarking

Accurate detection of landmark points plays an important role in many applications, such as face verification, face tracking, face expression analysis, 3D face modelling, etc. There are a lot of approaches proposed for this problem in the literature. These methods are generally based on two types of information: local texture around a given feature and geometric configuration of a given set of facial features or combination of two types of information. In this work, we investigate the performance of four transformation techniques: Gabor wavelets, independent component analysis (ICA), non-negative matrix factorization (NMF), discrete cosine transform (DCT). Since these methods are based on the texture information and aim to model the facial features in a suitable subspace, they are called ldquoappearance-based methodsrdquo. We also present some experimental results of proposed methods in conjunction with structural information.