Histogram of weighted local directions for gait recognition

In this paper, we explore the effectiveness of patch-based gradient feature extraction methods when applied to appearance-based gait recognition. Extending existing popular feature extraction methods such as HOG and LDP, we propose a novel technique which we term the Histogram of Weighted Local Directions (HWLD). These 3 methods are applied to gait recognition using the GEI feature, with classification performed using SRC. Evaluations on the CASIA and OULP datasets show significant improvements using these patch-based methods over existing implementations, with the proposed method achieving the highest recognition rate for the respective datasets. In addition, the HWLD can easily be extended to 3D, which we demonstrate using the GEV feature on the DGD dataset, observing improvements in performance.

Person re-identification using group information

After first observing a person, the task of person re-identification involves recognising an individual at different locations across a network of cameras at a later time. Traditionally, this task has been performed by first extracting appearance features of an individual and then matching these features to the previous observation. However, identifying an individual based solely on appearance can be ambiguous, particularly when people wear similar clothing (i.e. people dressed in uniforms in sporting and school settings). This task is made more difficult when the resolution of the input image is small as is typically the case in multi-camera networks. To circumvent these issues, we need to use other contextual cues. In this paper, we use "group" information as our contextual feature to aid in the re-identification of a person, which is heavily motivated by the fact that people generally move together as a collective group. To encode group context, we learn a linear mapping function to assign each person to a "role" or position within the group structure. We then combine the appearance and group context cues using a weighted summation. We demonstrate how this improves performance of person re-identification in a sports environment over appearance based-features.

Multi-shot person re-identification via relational stein divergence

Person re-identification is particularly challenging due to significant appearance changes across separate camera views. In order to re-identify people, a representative human signature should effectively handle differences in illumination, pose and camera parameters. While general appearance-based methods are modelled in Euclidean spaces, it has been argued that some applications in image and video analysis are better modelled via non-Euclidean manifold geometry. To this end, recent approaches represent images as covariance matrices, and interpret such matrices as points on Riemannian manifolds. As direct classification on such manifolds can be difficult, in this paper we propose to represent each manifold point as a vector of similarities to class representers, via a recently introduced form of Bregman matrix divergence known as the Stein divergence. This is followed by using a discriminative mapping of similarity vectors for final classification. The use of similarity vectors is in contrast to the traditional approach of embedding manifolds into tangent spaces, which can suffer from representing the manifold structure inaccurately. Comparative evaluations on benchmark ETHZ and iLIDS datasets for the person re-identification task show that the proposed approach obtains better performance than recent techniques such as Histogram Plus Epitome, Partial Least Squares, and Symmetry-Driven Accumulation of Local Features.

Inpatient facial expression recognition with SMS alert

Facial expression recognition (FER) has been dramatically developed in recent years, thanks to the advancements in related fields, especially machine learning, image processing and human recognition. Accordingly, the impact and potential usage of automatic FER have been growing in a wide range of applications, including human-computer interaction, robot control and driver state surveillance. However, to date, robust recognition of facial expressions from images and videos is still a challenging task due to the difficulty in accurately extracting the useful emotional features. These features are often represented in different forms, such as static, dynamic, point-based geometric or region-based appearance. Facial movement features, which include feature position and shape changes, are generally caused by the movements of facial elements and muscles during the course of emotional expression. The facial elements, especially key elements, will constantly change their positions when subjects are expressing emotions. As a consequence, the same feature in different images usually has different positions. In some cases, the shape of the feature may also be distorted due to the subtle facial muscle movements. Therefore, for any feature representing a certain emotion, the geometric-based position and appearance-based shape normally changes from one image to another image in image databases, as well as in videos. This kind of movement features represents a rich pool of both static and dynamic characteristics of expressions, which playa critical role for FER. The vast majority of the past work on FER does not take the dynamics of facial expressions into account. Some efforts have been made on capturing and utilizing facial movement features, and almost all of them are static based. These efforts try to adopt either geometric features of the tracked facial points, or appearance difference between holistic facial regions in consequent frames or texture and motion changes in loca- facial regions. Although achieved promising results, these approaches often require accurate location and tracking of facial points, which remains problematic.

Visual place recognition for persistent robot navigation in changing environments

This thesis demonstrates that robots can learn about how the world changes, and can use this information to recognise where they are, even when the appearance of the environment has changed a great deal. The ability to localise in highly dynamic environments using vision only is a key tool for achieving long-term, autonomous navigation in unstructured outdoor environments. The proposed learning algorithms are designed to be unsupervised, and can be generated by the robot online in response to its observations of the world, without requiring information from a human operator or other external source.

Unsupervised online learning of condition-invariant images for place recognition

This paper presents an online, unsupervised training algorithm enabling vision-based place recognition across a wide range of changing environmental conditions such as those caused by weather, seasons, and day-night cycles. The technique applies principal component analysis to distinguish between aspects of a location’s appearance that are condition-dependent and those that are condition-invariant. Removing the dimensions associated with environmental conditions produces condition-invariant images that can be used by appearance-based place recognition methods. This approach has a unique benefit – it requires training images from only one type of environmental condition, unlike existing data-driven methods that require training images with labelled frame correspondences from two or more environmental conditions. The method is applied to two benchmark variable condition datasets. Performance is equivalent or superior to the current state of the art despite the lesser training requirements, and is demonstrated to generalise to previously unseen locations.

Human mating in the informal market for sperm donation: Preferences and decision making

Choosing a mate is one of the largest (economic) decisions humans make. This thesis investigates this large scale decision and how the process is changing with the advent of the internet and the growing market for online informal sperm donation. This research identifies individual factors that influence female mating preferences. It explores the roles of behavioural traits and physical appearance, preferences for homogamy and hypergamy, and personality, and how these impact the decision to choose a donor. Overall, this thesis makes contributions to both the literature on human behaviour, and that on decision-making in extreme and highly important situations.

Integrating sorghum whole genome sequence information with a compendium of sorghum QTL studies reveals uneven distribution of QTL and of gene-rich regions with significant implications for crop improvement.

A comprehensive analysis was conducted using 48 sorghum QTL studies published from 1995 to 2010 to make information from historical sorghum QTL experiments available in a form that could be more readily used by sorghum researchers and plant breeders. In total, 771 QTL relating to 161 unique traits from 44 studies were projected onto a sorghum consensus map. Confidence intervals (CI) of QTL were estimated so that valid comparisons could be made between studies. The method accounted for the number of lines used and the phenotypic variation explained by individual QTL from each study. In addition, estimated centimorgan (cM) locations were calculated for the predicted sorghum gene models identified in Phytozome (JGI GeneModels SBI v1.4) and compared with QTL distribution genome-wide, both on genetic linkage (cM) and physical (base-pair/bp) map scales. QTL and genes were distributed unevenly across the genome. Heterochromatic enrichment for QTL was observed, with approximately 22% of QTL either entirely or partially located in the heterochromatic regions. Heterochromatic gene enrichment was also observed based on their predicted cM locations on the sorghum consensus map, due to suppressed recombination in heterochromatic regions, in contrast to the euchromatic gene enrichment observed on the physical, sequence-based map. The finding of high gene density in recombination-poor regions, coupled with the association with increased QTL density, has implications for the development of more efficient breeding systems in sorghum to better exploit heterosis. The projected QTL information described, combined with the physical locations of sorghum sequence-based markers and predicted gene models, provides sorghum researchers with a useful resource for more detailed analysis of traits and development of efficient marker-assisted breeding strategies.

Consumers' emotional judgement of materials durability and disposal

The literature demonstrates that understanding relating to the use of materials in product design has been investigated from both engineering and design perspectives. However, none of these studies have explored the consumers’ concepts of the materials; rather they have focused on participants’ discussions of material samples. Consumers’ emotional reactions to the materials themselves or the consumers’ reaction to the durability of the materials have not been previously explored in depth. This research has investigated these issues and has found that consumers have very specific concepts about materials. Furthermore, the combinations of consumer concepts that are likely to elicit an emotional judgement by the consumer have also been identified. It was found that consumers are conscious of the durability of their products and the materials that they are made from. This knowledge contributes to the support of environmentally conscious design, as well as user-centered design knowledge and practice. An understanding of the emotion consumers attribute to the effect wear and aging had on the materials’ physical appearance has been achieved. This understanding of consumers’ emotional reactions to materials can contribute not only to design considerations but to knowledge regarding the promotion of prolonged product-user relationships.

Assessing clinical reasoning: A method to monitor its development in a PBL curriculum

The aim of this study was to develop and trial a method to monitor the evolution of clinical reasoning in a PBL curriculum that is suitable for use in a large medical school. Termed Clinical Reasoning Problems (CRPs), it is based on the notion that clinical reasoning is dependent on the identification and correct interpretation of certain critical clinical features. Each problem consists of a clinical scenario comprising presentation, history and physical examination. Based on this information, subjects are asked to nominate the two most likely diagnoses and to list the clinical features that they considered in formulating their diagnoses, indicating whether these features supported or opposed the nominated diagnoses. Students at different levels of medical training completed a set of 10 CRPs as well as the Diagnostic Thinking Inventory, a self-reporting questionnaire designed to assess reasoning style. Responses were scored against those of a reference group of general practitioners. Results indicate that the CRPs are an easily administered, reliable and valid assessment of clinical reasoning, able to successfully monitor its development throughout medical training. Consequently, they can be employed to assess clinical reasoning skill in individual students and to evaluate the success of undergraduate medical schools in providing effective tuition in clinical reasoning.

Regulation of Reproduction in the Primitively Eusocial Wasp Ropalidia marginata: on the Trail of the Queen Pheromone

Queens and workers are not morphologically differentiated in the primitively eusocial wasp, Ropalidia marginata. Upon removal of the queen, one of the workers becomes extremely aggressive, but immediately drops her aggression if the queen is returned. If the queen is not returned, this hyper-aggressive individual, the potential queen (PQ), will develop her ovaries, lose her hyper-aggression, and become the next colony queen. Because of the non-aggressive nature of the queen, and because the PQ loses her aggression by the time she starts laying eggs, we hypothesized that regulation of worker reproduction in R marginata is mediated by pheromones rather than by physical aggression. Based on the immediate loss of aggression by the PQ upon return of the queen, we developed a bioassay to test whether the queen's Dufour's gland is, at least, one of the sources of the queen pheromone. Macerates of the queen's Dufour's gland, but not that of the worker's Dufour's gland, mimic the queen in making the PQ decrease her aggression. We also correctly distinguished queens and workers of R. marginata nests by a discriminant function analysis based on the chemical composition of their respective Dufour's glands.

Esclavos en el reino de Toledo

Resumen: La esclavitud en la Edad Moderna supuso una realidad social muy arraigada. Último eslabón de una sociedad estamental, en este trabajo, enmarcado espacialmente en el reino de Toledo en tierras de Castilla se analizan diversos aspectos de la condición de los esclavos: descripción y aspecto físico, tipología de las causas de esclavitud, comercio de esclavos, ocupaciones y servicios a que se dedicaban, comportamiento y conflictividad social y finalmente la manumisión. Para ello se han utilizado fuentes procedentes de archivos religiosos, de la nobleza y abundante documentación notarial y judicial.

Validade e confiabilidade do Questionário de Qualidade de Vida para Pacientes Orto-cirúrgicos (B-OQLQ)

O Orthognathic Quality of Life Questionnaire (OQLQ) foi desenvolvido em 2000 e validado em 2002, com o objetivo de analisar os impactos e benefícios do tratamento orto-cirúrgico na qualidade de vida dos pacientes. O objetivo deste trabalho foi avaliar a validade de construto e confiabilidade da versão brasileira deste instrumento, B-OQLQ, para verificar se, no processo de adaptação transcultural, as propriedades psicométricas do questionário original foram mantidas. Para tal, foi realizado um estudo seccional com uma amostra composta por cento e um pacientes com necessidade de tratamento ortodôntico-cirúrgico no Hospital Universitário Pedro Ernesto (HUPE) e na Faculdade de Odontologia da Universidade do Estado do Rio de Janeiro (UERJ), utilizando o B-OQLQ na forma de autopreenchimento. A média de idade dos pacientes foi de 26,51 (dp= 9,25), sendo a maioria do sexo feminino (58,42%; n=59) e maior que 21 anos (n=68, 67,33%). Quanto ao tipo de cirurgia, 42,57% (n=43) da amostra necessitavam fazer procedimento cirúrgico que envolveria as duas bases ósseas, 16,83% (n=17) necessitavam operar apenas uma das bases ósseas e para 40,59% (n=41) dos pacientes ainda não era possível definir qual o tipo de cirurgia seria necessária. Uma ortodontista treinada examinou os dentes dos pacientes para registrar o Índice de Dentes Cariados, Perdidos e Obturados (CPOD). A validade de construto foi acessada através do teste de correlação de Spearman entre as pontuações do B-OQLQ e do questionário Oral Health Impact Profile (OHIP-14) e das pontuações do B-OQLQ com as pontuações obtidas através de indicadores subjetivos e objetivos de saúde. A confiabilidade foi acessada em termos de consistência interna e estabilidade (teste-reteste), utilizando-se o alfa de Cronbach e o Coeficiente de Correlação Intraclasse (CCI), respectivamente. Os escores do B-OQLQ se correlacionaram significativamente com: o escore total do OHIP-14 (rs=0,70, p<0,001), a saúde bucal percebida (rs=-0,24, p=0,02), a qualidade de vida medida por um item único de avaliação (rs=-0,29, p=0,03), a satisfação com a aparência física (rs=-0,40, p<0,001) e a satisfação com a aparência facial (rs=-0,39, p=0,0001). Foi encontrada uma associação entre a faixa etária e o escore total do B-OQLQ (p=0,0012), sendo que pacientes maiores de 21 anos obtiveram escores mais elevados que pacientes mais novos. A associação entre o escore total do B-OQLQ e o tipo de cirurgia não foi estatisticamente significativa. O alfa de Cronbach e o CCI foram 0,95 e 0,90, respectivamente. Os domínios do B-OQLQ que mais causaram impacto na qualidade de vida foram aspectos sociais da deformidade (13,0; dp= 10,54) e estética facial (11,81; dp= 6,23). A versão brasileira do questionário OQLQ se mostrou um instrumento válido e confiável, com boas propriedades psicométricas podendo, portanto, ser considerada um instrumento apropriado para acessar o impacto da deformidade dentofacial na qualidade de vida de pacientes portadores desta condição.

Gure buruaren onarpenetik, gizakion errespetura

[eus] Gaur egungo eskoletan, oraindik ere, zoritxarrez, behar besteko garrantzirik ematen ez zaion eta sendotu behar den ikasleen autoezagutza fisikoaren inguruan, honako landa-lana diseinatu eta ikasleekin burutu egin da, hauengan aldaketa bat sortarazteko asmoarekin. Horretarako, prestatutako ariketa desberdinen bitartez, ikasleen arrazonamendu maila, baita maila emozionala ere landu egin da. Kontuan izanda, autoezagutzan, aldaketak epe luzean ematen direla, esku-hartze honen bidez ez dira aldaketa handirik edo esanguratsurik nabarmendu. Dena den, errespetuaren balorea ikasleengan transmititzea lortu egin da. Hori dela eta, landa lana burutu eta gero, egindakoaren inguruan, autokritika egin da.

Segmentation and recognition using structure from motion point clouds

We propose an algorithm for semantic segmentation based on 3D point clouds derived from ego-motion. We motivate five simple cues designed to model specific patterns of motion and 3D world structure that vary with object category. We introduce features that project the 3D cues back to the 2D image plane while modeling spatial layout and context. A randomized decision forest combines many such features to achieve a coherent 2D segmentation and recognize the object categories present. Our main contribution is to show how semantic segmentation is possible based solely on motion-derived 3D world structure. Our method works well on sparse, noisy point clouds, and unlike existing approaches, does not need appearance-based descriptors. Experiments were performed on a challenging new video database containing sequences filmed from a moving car in daylight and at dusk. The results confirm that indeed, accurate segmentation and recognition are possible using only motion and 3D world structure. Further, we show that the motion-derived information complements an existing state-of-the-art appearance-based method, improving both qualitative and quantitative performance. © 2008 Springer Berlin Heidelberg.