Illumination normalization for face recognition

Face recognition has been actively investigated by the scientific community and has already taken its place in modern consumer software. However, there are still major challenges remaining e.g. preventing negative influence from varying illumination, even with well known face recognition systems. To reduce the performance drop off caused by illumination, normalization methods can be applied as pre-processing step. Well known approaches are linear regression or local operations. In this publication the authors present the results of a two-step evaluation for real-world applications of a wide range of state-of-the-art illumination normalization algorithms. Further they present a new taxonomy of these methods and depict advanced algorithms motivated by the pre-eminent human abilities of illumination normalization. Additionally they introduce a recent bio-inspired algorithm based on diffusion filters that outperforms most of the known algorithms. Finally they deduce the theoretical potentials and practical applicability of the normalization methods from the evaluation results.