Illumination robust dictionary-based face recognition

TitleIllumination robust dictionary-based face recognition
Publication TypeConference Papers
Year of Publication2011
AuthorsPatel VM, Wu T, Biswas S, Phillips PJ, Chellappa R
Conference Name2011 18th IEEE International Conference on Image Processing (ICIP)
Date Published2011/09/11/14
ISBN Number978-1-4577-1304-0
Keywordsalbedo, approximation theory, classification, competitive face recognition algorithms, Databases, Dictionaries, Face, face recognition, face recognition method, filtering theory, human face recognition, illumination robust dictionary-based face recognition, illumination variation, image representation, learned dictionary, learning (artificial intelligence), lighting, lighting conditions, multiple images, nonstationary stochastic filter, publicly available databases, relighting, relighting approach, representation error, residual vectors, Robustness, simultaneous sparse approximations, simultaneous sparse signal representation, sparseness constraint, Training, varying illumination, vectors

In this paper, we present a face recognition method based on simultaneous sparse approximations under varying illumination. Our method consists of two main stages. In the first stage, a dictionary is learned for each face class based on given training examples which minimizes the representation error with a sparseness constraint. In the second stage, a test image is projected onto the span of the atoms in each learned dictionary. The resulting residual vectors are then used for classification. Furthermore, to handle changes in lighting conditions, we use a relighting approach based on a non-stationary stochastic filter to generate multiple images of the same person with different lighting. As a result, our algorithm has the ability to recognize human faces with good accuracy even when only a single or a very few images are provided for training. The effectiveness of the proposed method is demonstrated on publicly available databases and it is shown that this method is efficient and can perform significantly better than many competitive face recognition algorithms.