Cheung, Siu Ming. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2010. / Includes bibliographical references (leaves 81-86). / Abstracts in English and Chinese. / Chapter 1. --- INTRODUCTION --- p.1 / Chapter 1.1. --- Introduction to Face Recognition --- p.2 / Chapter 1.1.1. --- Modes of Face Recognition --- p.2 / Chapter 1.1.2. --- Typical Face Recognition System --- p.3 / Chapter 1.1.3. --- Face Recognition Algorithms --- p.4 / Chapter 1.1.4. --- The State of the Art --- p.5 / Chapter 1.2. --- Outdoor Face Recognition --- p.6 / Chapter 1.2.1. --- The Outdoor Environment --- p.6 / Chapter 1.2.2. --- The Illumination Variation Problem in the Outdoors --- p.8 / Chapter 1.3. --- Related works --- p.10 / Chapter 1.3.1. --- Face Appearance Modeling --- p.10 / Chapter 1.3.2. --- Illumination Invariant Features and Representations --- p.13 / Chapter 1.3.3. --- Active Near-Infrared Illumination --- p.14 / Chapter 1.4. --- Proposed method --- p.17 / Chapter 1.5. --- Design Requirements --- p.18 / Chapter 2. --- COMPENSATION METHODOLOGY FOR OUTDOOR FACE RECOGNITION --- p.20 / Chapter 2.1. --- Illumination from the Sun --- p.21 / Chapter 2.2. --- Effect of Sunlight Illumination --- p.22 / Chapter 2.3. --- A Compensation Model --- p.24 / Chapter 2.4. --- A Face Lighting Simulator --- p.28 / Chapter 2.4.1. --- Face 3D Models --- p.29 / Chapter 2.4.2. --- Light Sources --- p.30 / Chapter 2.4.3. --- Synthesis of Face Image --- p.31 / Chapter 2.5. --- Simulation Results --- p.32 / Chapter 2.5.1. --- Optimum Compensation Angles --- p.33 / Chapter 2.5.2. --- Effect of Illuminator Intensity --- p.36 / Chapter 2.5.3. --- Effect of Illuminator Elevation Angle --- p.38 / Chapter 2.5.4. --- Effect of Sunlight Elevation Angle --- p.41 / Chapter 2.5.5. --- Illumination from Both Sides --- p.42 / Chapter 2.6. --- Summary --- p.43 / Chapter 3. --- AN ADAPTIVE ILLUMINATOR --- p.45 / Chapter 3.1. --- Hardware Design --- p.45 / Chapter 3.1.1. --- Near-infrared Camera --- p.45 / Chapter 3.1.2. --- Illumination Panels --- p.48 / Chapter 3.1.3. --- Illuminator Controller --- p.56 / Chapter 3.1.4. --- Illumination Characteristics --- p.59 / Chapter 3.2. --- Algorithms --- p.62 / Chapter 3.2.1. --- Light Balance Estimation --- p.63 / Chapter 4. --- EXPERIMENTS AND RESULTS --- p.67 / Chapter 4.1. --- Effect of compensation angle on face similarity --- p.68 / Chapter 4.2. --- Effect of illumination compensation under different sunlight conditions --- p.71 / Chapter 4.3. --- Impact on recognition performance --- p.72 / Chapter 5. --- CONCLUSIONS --- p.76 / Chapter 6. --- BIBLIOGRAPHY --- p.81

Wu, King Keung. / "August 2011." / Thesis (M.Phil.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (p. 119-126). / Abstracts in English and Chinese. / Abstract --- p.i / Acknowledgement --- p.v / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Overview of Face Alignment Algorithms --- p.1 / Chapter 1.1.1 --- Objectives --- p.1 / Chapter 1.1.2 --- Motivation: Photo-realistic Talking Head --- p.2 / Chapter 1.1.3 --- Existing methods --- p.5 / Chapter 1.2 --- Contributions --- p.8 / Chapter 1.3 --- Outline of the Thesis --- p.11 / Chapter 2 --- Sparse Signal Representation --- p.13 / Chapter 2.1 --- Introduction --- p.13 / Chapter 2.2 --- Problem Formulation --- p.15 / Chapter 2.2.1 --- l0-nonn minimization --- p.15 / Chapter 2.2.2 --- Uniqueness --- p.16 / Chapter 2.3 --- Basis Pursuit --- p.18 / Chapter 2.3.1 --- From l0-norm to l1-norm --- p.19 / Chapter 2.3.2 --- l0-l1 Equivalence --- p.20 / Chapter 2.4 --- l1-Regularized Least Squares --- p.21 / Chapter 2.4.1 --- Noisy case --- p.22 / Chapter 2.4.2 --- Over-determined systems of linear equations --- p.22 / Chapter 2.5 --- Summary --- p.24 / Chapter 3 --- Sparse Corruptions and Principal Component Pursuit --- p.25 / Chapter 3.1 --- Introduction --- p.25 / Chapter 3.2 --- Sparse Corruptions --- p.26 / Chapter 3.2.1 --- Sparse Corruptions and l1-Error --- p.26 / Chapter 3.2.2 --- l1-Error and Least Absolute Deviations --- p.28 / Chapter 3.2.3 --- l1-Regularized l1-Error --- p.29 / Chapter 3.3 --- Robust Principal Component Analysis (RPCA) and Principal Component Pursuit --- p.31 / Chapter 3.3.1 --- Principal Component Analysis (PCA) and RPCA --- p.31 / Chapter 3.3.2 --- Principal Component Pursuit --- p.33 / Chapter 3.4 --- Experiments of Sparse and Low-rank Approach on Surveillance Video --- p.34 / Chapter 3.4.1 --- Least Squares --- p.35 / Chapter 3.4.2 --- l1-Regularized Least Squares --- p.35 / Chapter 3.4.3 --- l1-Error --- p.36 / Chapter 3.4.4 --- l1-Regularized l1-Error --- p.36 / Chapter 3.5 --- Summary --- p.37 / Chapter 4 --- Split Bregman Algorithm for l1-Problem --- p.45 / Chapter 4.1 --- Introduction --- p.45 / Chapter 4.2 --- Bregman Distance --- p.46 / Chapter 4.3 --- Bregman Iteration for Constrained Optimization --- p.47 / Chapter 4.4 --- Split Bregman Iteration for l1-Regularized Problem --- p.50 / Chapter 4.4.1 --- Formulation --- p.51 / Chapter 4.4.2 --- Advantages of Split Bregman Iteration . . --- p.52 / Chapter 4.5 --- Fast l1 Algorithms --- p.54 / Chapter 4.5.1 --- l1-Regularized Least Squares --- p.54 / Chapter 4.5.2 --- l1-Error --- p.55 / Chapter 4.5.3 --- l1-Regularized l1-Error --- p.57 / Chapter 4.6 --- Summary --- p.58 / Chapter 5 --- Face Alignment Using Sparse and Low-rank Decomposition --- p.61 / Chapter 5.1 --- Robust Alignment by Sparse and Low-rank Decomposition for Linearly Correlated Images (RASL) --- p.61 / Chapter 5.2 --- Problem Formulation --- p.62 / Chapter 5.2.1 --- Theory --- p.62 / Chapter 5.2.2 --- Algorithm --- p.64 / Chapter 5.3 --- Direct Extension of RASL: Multi-RASL --- p.66 / Chapter 5.3.1 --- Formulation --- p.66 / Chapter 5.3.2 --- Algorithm --- p.67 / Chapter 5.4 --- Matlab Implementation Details --- p.68 / Chapter 5.4.1 --- Preprocessing --- p.70 / Chapter 5.4.2 --- Transformation --- p.73 / Chapter 5.4.3 --- Jacobian Ji --- p.74 / Chapter 5.5 --- Experiments --- p.75 / Chapter 5.5.1 --- Qualitative Evaluations Using Small Dataset --- p.76 / Chapter 5.5.2 --- Large Dataset Test --- p.81 / Chapter 5.5.3 --- Conclusion --- p.85 / Chapter 5.6 --- Sensitivity analysis on selection of references --- p.87 / Chapter 5.6.1 --- References from consecutive frames --- p.88 / Chapter 5.6.2 --- References from RASL-aligned images --- p.91 / Chapter 5.7 --- Summary --- p.92 / Chapter 6 --- Extension of RASL for video: One-by-One Approach --- p.96 / Chapter 6.1 --- One-by-One Approach --- p.96 / Chapter 6.1.1 --- Motivation --- p.97 / Chapter 6.1.2 --- Algorithm --- p.97 / Chapter 6.2 --- Choices of Optimization --- p.101 / Chapter 6.2.1 --- l1-Regularized Least Squares --- p.101 / Chapter 6.2.2 --- l1-Error --- p.102 / Chapter 6.2.3 --- l1-Regularized l1-Error --- p.103 / Chapter 6.3 --- Experiments --- p.104 / Chapter 6.3.1 --- Evaluation for Different l1 Algorithms --- p.104 / Chapter 6.3.2 --- Conclusion --- p.108 / Chapter 6.4 --- Exploiting Property of Video --- p.109 / Chapter 6.5 --- Summary --- p.110 / Chapter 7 --- Conclusion and Future Work --- p.112 / Chapter A --- Appendix --- p.117 / Bibliography --- p.119

Although there is significant research surrounding online foreign language education, there are still questions on whether the outcomes are comparable to those obtained in the traditional face-to-face classroom. This study examined four classes, two online and two face-to-face, where students took the second course of a 6-level program of English as a Second Language at El Bosque University in Colombia. The International Test of English Proficiency (iTEP) was administered to students before classes started in order to establish a baseline, and then again after the courses finished. This test evaluates English language proficiency per skill: speaking, listening, reading, writing and also presents an overall proficiency score and level. Variables such as socioeconomic strata, students’ age, instructors, previous experience with online courses, course completion, student satisfaction and attendance and time on course were also examined. Because not all online students completed all content of the course within the timeframe given, special attention was given to this variable. Results indicate that when comparing the scores of only the online students who completed all the content of the course with those of their face-to-face counterparts, there are no statistically significant differences in the outcomes of any of the four skills nor there is a difference in the overall scores; however this brings up the issue of time investment as it seems to vary based upon instructional method. A qualitative component was used to support the findings in this study. This component included instructors’ interviews, an end-of-course qualitative survey and class observations.

Traditional economic theory suggests that consumers are likely to prepay for a product/ service that appears to be heavily discounted. However, in reality, many consumers do not think and act to achieve that goal. This is evident in consumer's psychological and behavioral responses toward a new type of price promotion, namely social coupons (SCs) (i.e., online coupons that offer consumers a substantial discount with a long redemption period when they prepay for a retailer's products/services). Such responses generate vital impacts not only on consumers themselves in terms of saving maximization but also on service retailers (e.g., sit-down dining restaurants) and SC providers (e.g., Groupon and LivingSocial) in terms of revenue maximization generated from offering SC campaigns. This dissertation aims to provide insights to the literature in price promotions, specifically SCs. Guided by mental accounting theory (i.e., consumers open a mental account when costs are incurred and close a mental account when benefits are received), this dissertation is structured in the form of two separate empirical essays. While Essay 1, "Prepaying Less is Preferable to Saving More: The Role of Pain of Prepayment Aversion in Social Coupon Purchasing Decision", focuses on opening a SC mental account (i.e., cost incurred), Essay 2, "Superfluous Spending: The Role of Neglected Mental Budget Depletion in Spending Decision when Redeeming Social Coupons", focuses on closing a SC mental account (i.e., benefit received). Essay 1 explored why consumers purchase SCs featuring a low-implausible face value (i.e., a face value that is lower than the normal price range expected by consumers for a particular type of service)? Findings across five experiments revealed that consumers' likelihood of purchasing SCs featuring a low-implausible (vs. plausible) face value was greater when a coupon price for SCs featuring a low-implausible face value was lower than willingness-to-prepay for a SC (WTPP-SC), while a coupon price for SCs featuring a plausible face value was higher than WTPP-SC. Furthermore, consumers' likelihood of purchasing SCs featuring a low-implausible face value was greater when a coupon price was lower (vs. higher) than WTPP-SC. Pain of prepayment (i.e., the disutility/imputed cost, painful feeling, generated from the thought of prepaying amount of money required for a SC) aversion was an underlying process. That is, consumers experienced greater pain of prepayment when a coupon price was higher (vs. lower) than WTPP-SC. Pain of prepayment, in turns, negatively influenced consumers' likelihood of purchasing SCs featuring a low-implausible face value. Moreover, consumers' likelihood of purchasing such SCs was greater when time pressure was present (vs. absent) and when semantic cues were abstract (vs. concrete). Finally, when being exposed to multiple SC deals for the same service, which vary in terms of face value plausibility (Option 1: low-implausible face value vs. Option 2: plausible face value), consumers were more likely to choose a SC deal featuring a low-implausible face value (Option 1) when a coupon price for a SC deal featuring a low-implausible face value was lower than WTPP-SC but a coupon price for a SC deal featuring a plausible face value (Option 2) was higher than WTPP-SC. In contrast, when coupon prices for both SC deal options were lower than WTPP-SC, consumers were more likely to choose a SC deal featuring a plausible face value (Option 2). Essay 2 explored why consumers spend a great additional amount of money beyond a SC face value? Findings across three experiments revealed that the amount of money spent beyond a SC face value was greater when consumers redeem SCs featuring a low-implausible (vs. plausible) face value. Neglected mental budget depletion (i.e., the instance in which consumers neglect the fact that the budget assigned to a particular SC mental account as a spending self-control is already depleted) was an underlying process. That is, consumers had a greater tendency to neglect mental budget depletion when redeeming SCs featuring a low-implausible (vs. plausible) face value. Neglected mental budget depletion, in turns, positively influenced the amount of money spent beyond a SC face value. Furthermore, concrete (vs. abstract) semantic cues and far (vs. near) distance between purchasing and redeeming a SC intensified neglected mental budget depletion effect, which in turns, increased the amount of money spent beyond a SC face value when redeeming SCs featuring a low-implausible face value. In conclusion, this dissertation provides theoretical insights on consumers' psychological responses, and their behavioral responses toward SCs during two SC stages, which results in sub-optimal SC decision-makings: (1) purchasing SCs featuring a low-implausible face value (Essay 1); and (2) spending additional money beyond a SC face value when redeeming SCs at a service retailer (Essay 2). The empirical findings across two essays add to the growing body of the literature in price promotions, specifically SCs. This dissertation also provides managerial insights regarding how managers can design and strategically implement SC campaigns that can maximize the number of SC being purchased and the great amount of money consumers spend beyond a SC face value when they redeem a SC at a service retailer.

The thesis argues that to find the places in media culture where the face transmits death, we need to look beyond the immobilised faces of the dead. Drawing on Walter Benjamin's philosophy of the image, the thesis sees the phenomenon of becoming unrecognisable as a particular practice of the image in which the face becomes a viable site for making death transmissible. It is argued that by paying attention to instances in media culture in which a face becomes unrecognisable, we can see how death is made visible as a dialectic between recognition and unrecognisability, appearance and disappearance. By examining the complexity of this particular form of dialectical image in a wide range of media - photography, television and film - the thesis shifts discussion away from questions of representation and faciality that feature so strongly in recent theorisations of the face. Focussing on questions of recognition and recognisability, the thesis proposes a way of thinking about the face that leads to a new conception of death in the media age / Doctor of Philosophy (PhD)

Cognitive impairment is a key characteristic of schizophrenia and is a clear predictor of functional outcome. This thesis explores the relationship between cognitive ability relating to social and non-social processing. Schizophrenia patients demonstrate an impaired ability to recognise, label and discriminate emotional expression within the face. The underlying mechanisms behind this social cognitive impairment are not yet fully understood. This thesis explores the notion that a basic perceptual impairment in processing facial information adversely impacts on the perception of more complex information derived from faces, such as emotional expression. Face perception relies on processing the featural characteristics of a face as well as the relationship between these features. Information pertaining to the spatial distances between features is referred to as configural information. / A group of schizophrenia patients and healthy control participants completed a battery of tasks that assessed basic neurocognition, facial emotion processing and configural face processing. A model of face processing was proposed and used to systematically pinpoint specific deficits that may contribute to impaired face processing in schizophrenia. The results indicated that schizophrenia patients show impairments on three broad constructs; basic neurocognition, facial emotion processing, and most pertinently, deficits in configural processing. It was revealed that although neurocognitive and face processing both explained a significant proportion of the variance in facial emotion processing, the effect of neurocognition was indirect and mediated by face processing. / To investigate the diagnostic specificity of these findings, a group of bipolar disorder patients was also tested on the task battery. The results indicated that bipolar disorder patients also show social and non-social cognitive impairments, however, not as severe as that demonstrated by the schizophrenia patients. Furthermore, the effect of neurocognitive performance on facial emotion processing appeared more direct for bipolar disorder patients compared to schizophrenia patients. Although deficits in face processing were observable in bipolar, they were not specific to configural processing. Thus, deficits in emotion processing were more associated to neurocognitive ability in bipolar disorder patients, and more associated to configural face processing in schizophrenia patients. The configural processing deficits in schizophrenia are discussed as a lower-order perception problem. In conclusion, the results of this thesis are discussed in terms of their implication for treatment.

Thesis (Ph. D.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

<p>Face detection is a classical application of object detection. There are many practical applications in which face detection is the first step; face recognition, video surveillance, image database management, video coding. </p><p>This report presents the results of an implementation of the AdaBoost algorithm to train a Strong Classifier to be used for face detection. The AdaBoost algorithm is fast and shows a low false detection rate, two characteristics which are important for face detection algorithms. </p><p>The application is an implementation of the AdaBoost algorithm with several command-line executables that support testing of the algorithm. The training and detection algorithms are separated from the rest of the application by a well defined interface to allow reuse as a software library. </p><p>The source code is documented using the JavaDoc-standard, and CppDoc is then used to produce detailed information on classes and relationships in html format. </p><p>The implemented algorithm is found to produce relatively high detection rate and low false alarm rate, considering the badly suited training data used.</p>

<p>Face detection has been a research area for more than ten years. It is a complex problem due to the high variability in faces and amongst faces; therefore it is not possible to extract a general pattern to be used for detection. This is what makes the face detection problem a challenge.</p><p>This thesis gives the reader a background to the face detection problem, where the two main approaches of the problem are described. A face detection algorithm is implemented using a context-based method in combination with an evolving neural network. The algorithm consists of two majors steps: detect possible face areas and within these areas detect faces. This method makes it possible to reduce the search space.</p><p>The performance of the algorithm is evaluated and analysed. There are several parameters that affect the performance; the feature extraction method, the classifier and the images used.</p><p>This work resulted in a face detection algorithm and the performance of the algorithm is evaluated and analysed. The analysis of the problems that occurred has provided a deeper understanding for the complexity of the face detection problem.</p>

The ability to detect faces in images is of critical ecological significance. It is a pre-requisite for other important face perception tasks such as person identification, gender classification and affect analysis. Here we address the question of how the visual system classifies images into face and non-face patterns. We focus on face detection in impoverished images, which allow us to explore information thresholds required for different levels of performance. Our experimental results provide lower bounds on image resolution needed for reliable discrimination between face and non-face patterns and help characterize the nature of facial representations used by the visual system under degraded viewing conditions. Specifically, they enable an evaluation of the contribution of luminance contrast, image orientation and local context on face-detection performance.