Face recognition algorithms has in the past few years become a very active area of research in the fields of computer vision, image processing, and cognitive psychology. This has spawned various algorithms of different complexities. The concept of principal component analysis(PCA) is a popular mode of face recognition algorithm and has often been used to benchmark other face recognition algorithms for identification and verification scenarios. However in this thesis, we try to analyze different face recognition algorithms at a deeper level. The objective is to model the distances output by any face recognition algorithm as a function of the input images. We achieve this by creating an affine eigen space from the PCA space such that it can approximate the results of the face recognition algorithm under consideration as closely as possible. Holistic template matching algorithms like the Linear Discriminant Analysis algorithm( LDA), the Bayesian Intrapersonal/Extrapersonal classifier(BIC), as well as local feature based algorithms like the Elastic Bunch Graph Matching algorithm(EBGM) and a commercial face recognition algorithm are selected for our experiments. We experiment on two different data sets, the FERET data set and the Notre Dame data set. The FERET data set consists of images of subjects with variation in both time and expression. The Notre Dame data set consists of images of subjects with variation in time. We train our affine approximation algorithm on 25 subjects and test with 300 subjects from the FERET data set and 415 subjects from the Notre Dame data set. We also analyze the effect of different distance metrics used by the face recognition algorithm on the accuracy of the approximation. We study the quality of the approximation in the context of recognition for the identification and verification scenarios, characterized by cumulative match score curves (CMC) and receiver operator curves (ROC), respectively. Our studies indicate that both the holistic template matching algorithms as well as feature based algorithms can be well approximated. We also find the affine approximation training can be generalized across covariates. For the data with time variation, we find that the rank order of approximation performance is BIC, LDA, EBGM, and commercial. For the data with expression variation, the rank order is LDA, BIC, commercial, and EBGM. Experiments to approximate PCA with distance measures other than Euclidean also performed very well. PCA+Euclidean distance is best approximated followed by PCA+MahL1, PCA+MahCosine, and PCA+Covariance.

In order to foster the maturity of face recognition analysis as a science, a well implemented baseline algorithm and good performance metrics are highly essential to benchmark progress. In the past, face recognition algorithms based on Principal Components Analysis(PCA) have often been used as a baseline algorithm. The objective of this thesis is to develop a strategy to estimate the best affine transformation, which when applied to the eigen space of the PCA face recognition algorithm can approximate the results of any given face recognition algorithm. The affine approximation strategy outputs an optimal affine transform that approximates the similarity matrix of the distances between a given set of faces generated by any given face recognition algorithm. The affine approximation strategy would help in comparing how close a face recognition algorithm is to the PCA based face recognition algorithm. This thesis work shows how the affine approximation algorithm can be used as a valuable tool to evaluate face recognition algorithms at a deep level. Two test algorithms were choosen to demonstrate the usefulness of the affine approximation strategy. They are the Linear Discriminant Analysis(LDA) based face recognition algorithm and the Bayesian interpersonal and intrapersonal classifier based face recognition algorithm. Our studies indicate that both the algorithms can be approximated well. These conclusions were arrived based on the results produced by analyzing the raw similarity scores and by studying the identification and verification performance of the algorithms. Two training scenarios were considered, one in which both the face recognition and the affine approximation algorithm were trained on the same data set and in the other, different data sets were used to train both the algorithms. Gross error measures like the average RMS error and Stress-1 error were used to directly compare the raw similarity scores. The histogram of the difference between the similarity matrixes also clearly showed that the error spread is small for the affine approximation algorithm. The performance of the algorithms in the identification and the verification scenario were characterized using traditional CMS and ROC curves. The McNemar's test showed that the difference between the CMS and the ROC curves generated by the test face recognition algorithms and the affine approximation strategy is not statistically significant. The results were statistically insignificant at rank 1 for the first training scenario but for the second training scenario they became insignificant only at higher ranks. This difference in performance can be attributed to the different training sets used in the second training scenario.

Soit k un corps quelconque. Dans cette th±se, on étudie le groupe de Picard des k-groupes algébriques unipotents (lisses et connexes).Tout k-groupe algébrique unipotent est extension itérée de formes du groupe additif; on va donc d'abord s'intéresser au groupe de Picard des formes du groupe additif. L'étude de ce groupe est faite avec une méthode géométrique qui permet de traiter le cas plus général des formes de la droite affine. On obtient ainsi une borne explicite sur la torsion du groupe de Picard desformes de la droite affine et sur la torsion de la composante neutre du foncteur de Picard de leur complétion régulière. De plus, on trouve une condition suffisante pour que le groupe de Picard d'une forme de la droite affinesoit non trivial et on construit des exemples de formes non triviales de la droite affine dont le groupe de Picard est trivial.Un k-groupe algébrique unipotent est une forme de l'espace affine. Afin d'étudier le groupe de Picard d'une forme X de l'espace affine avec une méthode géométrique, on définit un foncteur de Picard "restreint". On montre que si X admet une complétion régulière, alors le foncteur de Picard "restreint" est représentable par un k-groupe unipotent (lisse, non nécessairement connexe).Avec ce foncteur de Picard "restreint" et des raisonnements purement géométriques, on obtient que le groupe de Picard d'une forme unirationnelle de l'espace affine est fini. De plus, on généralise un résultat dû à B. Totaro: si k est séparablement clos, et si le groupe de Picard d'un k-groupe algébrique unipotent commutatif est non trivial, alors il admet une extension non triviale par le groupe multiplicatif. / Let k be any field. In this Ph.D. dissertation we study the Picard group of the (smooth connected) unipotent k-algebraic groups.As every unipotent algebraic group is an iterated extension of forms of the additive group, we will study the Picard group of the forms of the additive group. In fact we study the Picard group of forms of the additive group and the affine line simultaneously using a geometric method. We obtain anexplicit upper bound on the torsion of the Picard group of the forms of the affine line and their regular completion, and a sufficient condition for the Picard group of a form of the affine line to be nontrivial. We also give examples of nontrivial forms of the affine line with trivial Picard groups.In general, a unipotent k-algebraic group is a form of the affine n-space. In order to study the Picard group of a form X of the affine n-space with a geometric method, we define a "restricted" Picard functor; we show that if X admits a regular completion then the "restricted" Picard functor is representable by a unipotent k-algebraic group (smooth, not necessarly connected). With this "restricted" Picard functor and geometric arguments we show that the Picard group of a unirational form of the affine n-space is finite. Moreover we generalise a result of B. Totaro: if k is separablyclosed and if the Picard group of a unipotent k-algebraic group is nontrivial then it admits a nontrivial extension by the multiplicative group.