In Modellen der Gesichtererkennung (Bruce & Young, 1986) wird eine Verarbeitungsstufe postuliert, auf welcher strukturelle Gesichterrepraesentationen im Langzeitgedaechtnis (LZG) innerhalb von sogenannten Face Recognition Units (FRUs), abgerufen werden. Frühere Studien (z.B. Pfütze, Sommer & Schweinberger, 2002) zeigten eine Komponente in den ereigniskorrelierten Potentialen (EKPs), welche die Aktivierung solcher FRUs anzeigen - der frühe Wiederholungseffekt (oder early repetition effect). Dieser zeigt sich bei wiederholter Darbietung bekannter Gesichter um 250-350 ms mit frontaler Positivierung und temporal inferiorer Negativierung als ein EKP auf die zweite Praesentation. In Experiment I und II dieser Dissertation wurde der ERE als Marker der Aktivierung struktureller Gesichterrepraesentationen im LZG durch ein Prime-Target-Paradigma mit einer perzeptuellen Rueckwaertsmaskierung validiert. Die Maskierung mit einem unbekannten Gesicht ergab im Vergleich zu einem zerwürfelten Gesicht und der unmaskierten Bedingung eine spezifische Topographie des ERE, die den gedächtnisbasierten Anteil des ERE widerspiegelt. Dieses Paradigma wurde in Experiment III eingesetzt, um die Blickwinkelabhängigkeit des ERE zu untersuchen. Die Ergebnisse deuten darauf hin, daß Gesichterdurch mehr als nur blickwinkelabhängige, zweidimensionale Abbilder repräsentiert sind, da zumindest für Frontalgesichter ein ERE sogar über eine Blickwinkelabweichung von 90 Grad gezeigt werden konnte. Dies ist vereinbar mit einer "dualen Strategie" der Gesichtererkennung (z.B. O''Toole, Edelman & Bülthoff, 1998) wobei sowohl Konfiguration als auch Einzelmerkmale eines Gesichts abgerufen werden. Die Ergebnisse sind vereinbar mit Bruce und Young (1986), welche FRUs als Einheiten konzipierten, in denen konfigurale und einzelne Merkmale verschiedener Blickwinkel miteinander verbunden sind, was in diesem Sinn einer objektzentrierten, dreidimensionalen Speicherung eines individuellen Gesichts entspricht. / Models of face recognition (Bruce & Young, 1986) postulate a processing stage where face recognition units (FRUs) are accessed. FRUs are thought to be structural representations of familiar faces stored in long-term memory (LTM). Previous work (e.g. Pfütze, Sommer & Schweinberger, 2002), suggested the existence of a component in the event-related-potential (ERP) which signals FRU activation - the early repetition effect (ERE). The ERE can be observed when familiar faces are shown repeatedly, as an ERP around 250-350 ms that is more positive at fronto-central and more negative at inferior temporal sites for the second presentation of the face. In Experiment I and II of the present dissertation the ERE was validated as a marker of FRU activation in LTM by using a repetition priming paradigm with backward masking. Compared to the scrambled and non-mask conditions, the ERE revealed a distinguishable topography when an unfamiliar face interspersed prime and target, demonstrating contributions of face identity codes to the ERE. In Experiment III, this paradigm was used to analyse the viewpoint-dependency of the ERE. Results of Experiment III indicate that facial representations are more than viewpoint-dependent, two-dimensional images of a face, because even under a 90° view deviance between prime and target an ERE was detectable at least for frontal targets. This reconciles a dual-strategy in face recognition (e.g. O’Toole, Edelman & Bülthoff, 1998). Accordingly, facial representations can be assumed as a combination of image-based views, reflecting their configuration and single features. Results are in line with Bruce and Young (1986) who postulated that FRUs contain both features and configurations of distinct head angles in an interlinked manner. In that sense, FRUs can be described as three-dimensional analogues of the face they represent making object-centered recognition on an individual level possible.

In this thesis, the expression variation problem in two-dimensional (2D) and three-dimensional (3D) face recognition is tackled. While discriminant analysis (DA) methods are effective solutions for recognizing expression-variant 2D face images, they are not directly applicable when only a single sample image per subject is available. This problem is addressed in this thesis by introducing expression subspaces which can be used for synthesizing new expression images from subjects with only one sample image. It is proposed that by augmenting a generic training set with the gallery and their synthesized new expression images, and then training DA methods using this new set, the face recognition performance can be significantly improved. An important advantage of the proposed method is its simplicity; the expression of an image is transformed simply by projecting it into another subspace. The above proposed solution can also be used in general pattern recognition applications. The above method can also be used in 3D face recognition where expression variation is a more serious issue. However, DA methods cannot be readily applied to 3D faces because of the lack of a proper alignment method for 3D faces. To solve this issue, a method is proposed for sampling the points of the face that correspond to the same facial features across all faces, denoted as the closest-normal points (CNPs). It is shown that the performance of the linear discriminant analysis (LDA) method, applied to such an aligned representation of 3D faces, is significantly better than the performance of the state-of-the-art methods which, rely on one-by-one registration of the probe faces to every gallery face. Furthermore, as an important finding, it is shown that the surface normal vectors of the face provide a higher level of discriminatory information rather than the coordinates of the points. In addition, the expression subspace approach is used for the recognition of 3D faces from single sample. By constructing expression subspaces from the surface normal vectors at the CNPs, the surface normal vectors of a 3D face with single sample can be synthesized under other expressions. As a result, by improving the estimation of the within-class scatter matrix using the synthesized samples, a significant improvement in the recognition performance is achieved.

In this thesis, the expression variation problem in two-dimensional (2D) and three-dimensional (3D) face recognition is tackled. While discriminant analysis (DA) methods are effective solutions for recognizing expression-variant 2D face images, they are not directly applicable when only a single sample image per subject is available. This problem is addressed in this thesis by introducing expression subspaces which can be used for synthesizing new expression images from subjects with only one sample image. It is proposed that by augmenting a generic training set with the gallery and their synthesized new expression images, and then training DA methods using this new set, the face recognition performance can be significantly improved. An important advantage of the proposed method is its simplicity; the expression of an image is transformed simply by projecting it into another subspace. The above proposed solution can also be used in general pattern recognition applications. The above method can also be used in 3D face recognition where expression variation is a more serious issue. However, DA methods cannot be readily applied to 3D faces because of the lack of a proper alignment method for 3D faces. To solve this issue, a method is proposed for sampling the points of the face that correspond to the same facial features across all faces, denoted as the closest-normal points (CNPs). It is shown that the performance of the linear discriminant analysis (LDA) method, applied to such an aligned representation of 3D faces, is significantly better than the performance of the state-of-the-art methods which, rely on one-by-one registration of the probe faces to every gallery face. Furthermore, as an important finding, it is shown that the surface normal vectors of the face provide a higher level of discriminatory information rather than the coordinates of the points. In addition, the expression subspace approach is used for the recognition of 3D faces from single sample. By constructing expression subspaces from the surface normal vectors at the CNPs, the surface normal vectors of a 3D face with single sample can be synthesized under other expressions. As a result, by improving the estimation of the within-class scatter matrix using the synthesized samples, a significant improvement in the recognition performance is achieved.

Face recognition finds various applications in surveillance, Law enforcement etc. These applications require fast image processing in real time. Modern GPUs have evolved fully programmable parallel stream processors. The problem of face recognition in real time system is benefited by parallelism. With the aim of fulfilling both speed and accuracy criteria we present a GPU accelerated Face Recognition system. OpenCL is a heterogeneous computing language that allows extracting parallelism on different platforms like DSP processors, FPGAs, GPUs. The proposed kernel on GPU exploits coarse grain parallelism for Local Binary Pattern (LBP) histogram computation and ELTP (Enhanced Local Ternary Pattern) feature extraction. The proposed optimization techniques on local memory, work group size and work group dimension enhances the computation of face recognition on GPU further. As a result, we have achieved a speed up of 30 times to 300 times for 124*124 to 2048*2048 image sizes for LBP and ELTP feature extraction compared to CPU. We also present a robust real time face recognition and tracking on GPU using fusion of RGB and Depth images taken from Kinect sensor. The proposed segmentation after detection algorithm enhances the performances of recognition using LBP.

Syftet med rapporten är att visa möjligheten att använda Raspberry Pi 2 i ett ansiktsigenkänningssystem. Studien redogör för prestandaskillnader mellan Eigenface och Fisherfacemetoden. Studieförfattarna har genomfört en experimentell studie enligt en kvantitativ metod där tester utgör empirin. Resultatet från testerna kommer presenteras genom diagram och påvisa möjligheten att använda Raspberry Pi 2 som hårdvara i ett ansiktsigenkänningssystem. Genom samma testutförande kommer skillnader mellan igenkänningsmetoderna att påvisas. Studien visar att Raspberry Pi 2 är en lämplig kandidat att använda för mindre ansiktsigenkänningssystem. Vidare framgår det att Fisherface-metoden är det lämpligaste valet att använda vid implementation av systemet. / The purpose with this report is to demonstrate the possibility to use Raspberry Pi 2 as hardware in a face recognition system. The study will show performance differences regarding the Eigenface- and Fisherface-method. To demonstrate the possibility the authors have done tests using an experimental study and quantitative method. To review the tests and to understand the result a qualitative literature review was taken. The tests will be presented as graphs to show the possibility to use Raspberry Pi 2 as hardware in a face recognition system. The same goes for the comparison of the chosen algorithms. The work indicates that Raspberry Pi 2 is a possible candidate to use for smaller face recognition systems. There is also an indication that the Fisherface method is the better choice for face recognition.

Thesis (PhD)--University of Stellenbosch, 2001. / ENGLISH ABSTRACT: In this thesis we examine some aspects of automatic face recognition, with specific reference to the eigenface technique. We provide a thorough theoretical analysis of this technique which allows us to explain many of the results reported in the literature. It also suggests that clustering can improve the performance of the system and we provide experimental evidence of this. From the analysis, we also derive an efficient algorithm for updating the eigenfaces. We demonstrate the ability of an eigenface-based system to represent faces efficiently (using at most forty values in our experiments) and also demonstrate our updating algorithm. Since we are concerned with aspects of face recognition, one of the important practical problems is locating the face in a image, subject to distortions such as rotation. We review two well-known methods for locating faces based on the eigenface technique.e These algorithms are computationally expensive, so we illustrate how the Synthetic Discriminant Function can be used to reduce the cost. For our purposes, we propose the concept of a linearly interpolating SDF and we show how this can be used not only to locate the face, but also to estimate the extent of the distortion. We derive conditions which will ensure a SDF is linearly interpolating. We show how many of the more popular SDF-type filters are related to the classic SDF and thus extend our analysis to a wide range of SDF-type filters. Our analysis suggests that by carefully choosing the training set to satisfy our condition, we can significantly reduce the size of the training set required. This is demonstrated by using the equidistributing principle to design a suitable training set for the SDF. All this is illustrated with several examples. Our results with the SDF allow us to construct a two-stage algorithm for locating faces. We use the SDF-type filters to obtain initial estimates of the location and extent of the distortion. This information is then used by one of the more accurate eigenface-based techniques to obtain the final location from a reduced search space. This significantly reduces the computational cost of the process. / AFRIKAANSE OPSOMMING: In hierdie tesis ondersoek ons sommige aspekte van automatiese gesigs- herkenning met spesifieke verwysing na die sogenaamde eigengesig ("eigen- face") tegniek. ‘n Deeglike teoretiese analise van hierdie tegniek stel ons in staat om heelparty van die resultate wat in die literatuur verskyn te verduidelik. Dit bied ook die moontlikheid dat die gedrag van die stelsel sal verbeter as die gesigte in verskillende klasse gegroepeer word. Uit die analise, herlei ons ook ‘n doeltreffende algoritme om die eigegesigte op te dateer. Ons demonstreer die vermoë van die stelsel om gesigte op ‘n doeltreffende manier te beskryf (ons gebruik hoogstens veertig eigegesigte) asook ons opdateringsalgoritme met praktiese voorbeelde. Verder ondersoek ons die belangrike probleem om gesigte in ‘n beeld te vind, veral as rotasie- en skaalveranderinge plaasvind. Ons bespreek twee welbekende algoritmes om gesigte te vind wat op eigengesigte gebaseer is. Hierdie algoritme is baie duur in terme van numerise berekeninge en ons ontwikkel n koste-effektiewe metode wat op die sogenaamde "Synthetic Discriminant Functions" (SDF) gebaseer is. Vir hierdie doel word die begrip van lineêr interpolerende SDF’s ingevoer. Dit stel ons in staat om nie net die gesig te vind nie, maar ook ‘n skatting van sy versteuring te bereken. Voorts kon ons voorwaardes aflei wat verseker dat ‘n SDF lineêr interpolerend is. Aangesien ons aantoon dat baie van die gewilde SDF-tipe filters aan die klassieke SDF verwant is, geld ons resultate vir ‘n hele verskeidenheid SDF- tipe filters. Ons analise toon ook dat ‘n versigtige keuse van die afrigdata mens in staat stel om die grootte van die afrigstel aansienlik te verminder. Dit word duidelik met behulp van die sogenaamde gelykverspreidings beginsel ("equidistributing principle") gedemonstreer. Al hierdie aspekte van die SDF’s word met voorbeelde geïllustreer. Ons resultate met die SDF laat ons toe om ‘n tweestap algoritme vir die vind van ‘n gesig in ‘n beeld te ontwikkel. Ons gebruik eers die SDF-tipe filters om skattings vir die posisie en versteuring van die gesig te kry en dan verfyn ons hierdie skattings deur een van die teknieke wat op eigengesigte gebaseer is te gebruik. Dit lei tot ‘n aansienlike vermindering in die berekeningstyd.

This thesis relates to the design, implementation and evaluation of statistical face recognition techniques. In particular, the use of Hidden Markov Models in various forms is investigated as a recognition tool and critically evaluated. Current face recognition techniques are very dependent on issues like background noise, lighting and position of key features (ie. the eyes, lips etc.). Using an approach which specifically uses an embedded Hidden Markov Model along with spectral domain feature extraction techniques, shows that these dependencies may be lessened while high recognition rates are maintained.

Dual-process models of recognition memory propose that recognition memory can be supported by either a general sense of familiarity or the recollection of the encoding context. One source of evidence supporting dual-process models comes from event-related potential (ERP) studies of recognition memory, which have identified distinct patterns of neural activity associated with familiarity and recollection (the mid frontal and left parietal old/new effects, respectively). In this thesis, dual-process accounts of recognition memory were investigated in a series of ERP studies using three categories of stimulus: previously unknown faces, famous faces, and names. For previously unknown faces, familiarity was associated with activity over posterior scalp electrodes while recollection was associated with topographically dissociable activity over anterior electrodes. These dissociable patterns of activity support dual-process models. However, the typical pattern of old/new effects was only observed for stimuli associated with pre-existing representations (i.e., names and famous faces), suggesting that the presence/absence of pre-existing representations may determine the particular retrieval processes that support recognition memory. Furthermore, recollection was associated with two different patterns of activity (anterior and left parietal effects), suggesting that recollection is not a homogenous process. Dual-process theories may represent an important starting point for investigating recognition memory, but neither familiarity nor recollection appear to be functionally homogenous processes when theorizing is constrained by the analysis of scalp recorded electrophysiological activity.

Le but de l’expérience décrite dans ce mémoire est d'arriver à inculquer inconsciemment aux sujets une stratégie visuelle leur permettant d'utiliser seulement une partie spécifique de l'information visuelle disponible dans le visage humain pour en reconnaître le genre. Normalement, le genre d’un visage est reconnu au moyen de certaines régions, comme la bouche et les yeux (Dupuis-Roy, Fortin, Fiset et Gosselin, 2009). La tâche accomplie par les sujets permettait un apprentissage perceptuel implicite qui se faisait par conditionnement opérant. Ces derniers étaient informés qu'un nombre de points leur serait attribué selon leur performance à la tâche. Au terme de l’entraînement, les sujets renforcés pour l’utilisation de l’oeil gauche utilisaient davantage l’oeil gauche que l’oeil droit et ceux renforcés pour l’utilisation de l’oeil droit utilisaient davantage l’oeil droit. Nous discuterons de potentielles applications cliniques de cette procédure de conditionnement. / The goal of the following experiment is to make subjects unconsciously learn a visual strategy allowing them to use only part of the available visual information from the human face to correctly identify the gender of a face. Normally, the gender of a face is recognized using certain regions, like those of the mouth and the eyes (Dupuis-Roy, Fortin, Fiset et Gosselin, 2009). Our participants had to accomplish an operant conditionning task. They were informed that a number of points would be given to them according to their performance. At the end of training, the subjects that were encouraged to use the left eye indeed used the left eye more than the right. Also, those that were conditionned to use the right eye used the right eye more than the left. We will discuss the potential clinical applications of this method of conditionning.

The aim of the diploma thesis was to observe the capability of imitation in orofacial area of three days old infants and analyze if the capability relates to quality of psychomotor development, thus whether it relates with a global motor model. This paper is a pilot study comprising seventeen infants. Each of them was supposed to attend four visits during the first six months of their life. It included visits in the first few days, between the fourth and the sixth week, the third month and the sixth month of their life. The psychomotor development was examined and within the first visit there was established communication with the child by means of eye contact and imitation. Each child's attempt at imitation and response was recorded. This thesis can serve as a basis for further research dealing with this topic that includes the imitation skill of mirror neurons related to psychomotor development.