Understanding how the human visual system recognizes objects is one of the key challenges in neuroscience. Inspired by a large body of physiological evidence (Felleman and Van Essen, 1991; Hubel and Wiesel, 1962; Livingstone and Hubel, 1988; Tso et al., 2001; Zeki, 1993), a general class of recognition models has emerged which is based on a hierarchical organization of visual processing, with succeeding stages being sensitive to image features of increasing complexity (Hummel and Biederman, 1992; Riesenhuber and Poggio, 1999; Selfridge, 1959). However, these models appear to be incompatible with some well-known psychophysical results. Prominent among these are experiments investigating recognition impairments caused by vertical inversion of images, especially those of faces. It has been reported that faces that differ "featurally" are much easier to distinguish when inverted than those that differ "configurally" (Freire et al., 2000; Le Grand et al., 2001; Mondloch et al., 2002) ??finding that is difficult to reconcile with the aforementioned models. Here we show that after controlling for subjects' expectations, there is no difference between "featurally" and "configurally" transformed faces in terms of inversion effect. This result reinforces the plausibility of simple hierarchical models of object representation and recognition in cortex.

This thesis researches techniques for modelling static facial expressions, as well as the dynamics of continuous facial motion. We demonstrate how static and dynamic properties of facial expressions can be represented within a linear and nonlinear context, respectively. These two representations do not act in isolation, but are mutually reinforcing in conceding a cohesive framework for the analysis, animation, and manipulation of expressive faces. We derive a basis for the linear space of expressions through Principal Components Analysis (PCA). We introduce and formalize the notion of "expression manifolds", manifolds residing in PCA space that model motion dynamics for semantically similar expressions. We then integrate these manifolds into an animation workflow by performing Nonlinear Dimensionality Reduction (NLDR) on the expression manifolds. This operation yields expression maps that encode a wealth of information relating to complex facial dynamics, in a low dimensional space that is intuitive to navigate and efficient to manage.

This thesis researches techniques for modelling static facial expressions, as well as the dynamics of continuous facial motion. We demonstrate how static and dynamic properties of facial expressions can be represented within a linear and nonlinear context, respectively. These two representations do not act in isolation, but are mutually reinforcing in conceding a cohesive framework for the analysis, animation, and manipulation of expressive faces. We derive a basis for the linear space of expressions through Principal Components Analysis (PCA). We introduce and formalize the notion of "expression manifolds", manifolds residing in PCA space that model motion dynamics for semantically similar expressions. We then integrate these manifolds into an animation workflow by performing Nonlinear Dimensionality Reduction (NLDR) on the expression manifolds. This operation yields expression maps that encode a wealth of information relating to complex facial dynamics, in a low dimensional space that is intuitive to navigate and efficient to manage.

No description available.

<p>Any information represented in the brain, whether an individual is aware of it or not, holds the potential to affect behavior. The extent of visual perceptual processing that occurs in the absence of awareness is therefore a question of broad import and interest to the field of cognitive neuroscience. A useful approach for examining the extent and quality of visual processing that occurs in the absence of awareness is the dissociation paradigm. In this approach, experimenters track implicit measures of the visual process of interest across conditions of awareness modulated by visual presentation manipulations. Object-category discrimination by the visual system represents a relatively sophisticated level of representation that may or may not occur in the absence of awareness. Here, electrophysiological measures (scalp-recorded event-related potentials, or ERPs) of object-category discrimination by the brain (the face-specific N170 ERP component and the longer-latency face-specific negativity) were tracked across conditions of visual awareness as manipulated by multiple presentation paradigms (sandwich masking, object-substitution masking, the attentional blink, and motion-induced blindness). In addition, where possible, other related comparisons examining lower-level visual processes and higher-level attentional processes were employed to help delineate the specific level and mechanism by which awareness was disrupted in each case. The experiments implicated a unique set of mechanisms of reducing awareness for each method, while providing insight into the complex relationships between the various phases of visual processing in the human brain and awareness. Ultimately it was observed that neural indices of face-specific processing are differentially susceptible to disruption exerted by these various methods, and that there do in fact exist conditions in which awareness can be disrupted while leaving various facets and phases of face-specific processing intact. These findings help to establish object-category discrimination as a process that can occur in the absence of visual awareness, and contributes to our understanding of the neural factors that influence and determine behavior.</p> / Dissertation

There is a paucity of research examining the neurobiological functioning of new mothers who have experienced parental loss during development. The current study investigated the relationship between inconsistent (IC) versus consistent (CC) care and brain activity in regions that comprise a putative neurobiological model of mothering. Mothers were shown positive and negative pictures of their own and an unfamiliar infant. Through repeated measures ANOVAs, it was found that BOLD activity was greater for own infant in the nucleus accumbens (NAC) and amygdala (AMY) and that positive pictures elicited greater BOLD response in the NAC, AMY and anterior cingulate cortex. Interestingly, IC mothers show an even greater response own infant in the NAC and left hypothalamus (HYPO). In the left dorsolateral prefrontal cortex, IC mothers showed greater BOLD response to other infant. Thus, functioning of the maternal circuit, which includes areas strongly implicated in reward, may be altered by early experiences.

There is a paucity of research examining the neurobiological functioning of new mothers who have experienced parental loss during development. The current study investigated the relationship between inconsistent (IC) versus consistent (CC) care and brain activity in regions that comprise a putative neurobiological model of mothering. Mothers were shown positive and negative pictures of their own and an unfamiliar infant. Through repeated measures ANOVAs, it was found that BOLD activity was greater for own infant in the nucleus accumbens (NAC) and amygdala (AMY) and that positive pictures elicited greater BOLD response in the NAC, AMY and anterior cingulate cortex. Interestingly, IC mothers show an even greater response own infant in the NAC and left hypothalamus (HYPO). In the left dorsolateral prefrontal cortex, IC mothers showed greater BOLD response to other infant. Thus, functioning of the maternal circuit, which includes areas strongly implicated in reward, may be altered by early experiences.

Understanding how the human visual system recognizes objects is one of the key challenges in neuroscience. Inspired by a large body of physiological evidence (Felleman and Van Essen, 1991; Hubel and Wiesel, 1962; Livingstone and Hubel, 1988; Tso et al., 2001; Zeki, 1993), a general class of recognition models has emerged which is based on a hierarchical organization of visual processing, with succeeding stages being sensitive to image features of increasing complexity (Hummel and Biederman, 1992; Riesenhuber and Poggio, 1999; Selfridge, 1959). However, these models appear to be incompatible with some well-known psychophysical results. Prominent among these are experiments investigating recognition impairments caused by vertical inversion of images, especially those of faces. It has been reported that faces that differ “featurally” are much easier to distinguish when inverted than those that differ “configurally” (Freire et al., 2000; Le Grand et al., 2001; Mondloch et al., 2002) – a finding that is difficult to reconcile with the aforementioned models. Here we show that after controlling for subjects’ expectations, there is no difference between “featurally” and “configurally” transformed faces in terms of inversion effect. This result reinforces the plausibility of simple hierarchical models of object representation and recognition in cortex.

Made available in DSpace on 2014-12-17T15:37:00Z (GMT). No. of bitstreams: 1 TiagoJBE_DISSERT.pdf: 669719 bytes, checksum: 7259246da23dc724215353b961ad242c (MD5) Previous issue date: 2010-08-19 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / Todas as sociedades apresentam uma ordem, esta caracterizada por um conjunto de normas sobre o que ? permitido e o que deve ser coibido. A compreens?o sobre a origem, fixa??o e evolu??o das normas de conviv?ncia ? fundamental para entender a sociedade tal como ela ?, bem como o real papel dos indiv?duos que as comp?em. Escrevemos um artigo te?rico abordando essa id?ia. Apresentamos a Teoria do Contrato Social como um modelo para o estudo da evolu??o dos mecanismos cognitivos que regulam as trocas sociais. Esta teoria postula a exist?ncia de um m?dulo espec?fico de trapa?a. Diversos trabalhos procuraram testar essa hip?tese, inclusive relacionando essa adapta??o com a melhor recorda??o de faces de trapaceiros. No entanto, observamos que estes apresentam resultados divergentes sobre a exist?ncia ou n?o de um suposto m?dulo para a detec??o de trapaceiros. Al?m disso, foi poss?vel constatar falhas metodol?gicas nos trabalhos anteriores e a neglig?ncia no estudo da rela??o entre a lembran?a e vari?veis inerentes a qualquer indiv?duo como o sexo e diferen?as quanto ? tonalidade da cor da pele. Assim, se fez necess?ria uma nova investiga??o, descrita no artigo emp?rico. Para tanto, expusemos 190 alunos do ensino m?dio e universit?rio a oito v?deos de 30 cada, aonde era exibidas cenas de um sujeito est?mulo (SE; 4 homens e 4 mulheres) com diferentes tonalidades de cor de pele. Fazendo uso de fones de ouvido os participantes ouviam uma narra??o, a qual identificava o SE como n?o confi?vel ou confi?vel. Depois de uma tarefa de distra??o, os participantes assistiam a uma sess?o de fotos de fotos com 34 SE, incluindo os 8 SE vistos nos v?deos. Os participantes eram solicitados para apontar quais SE foram exibidos nos v?deos. Nossos resultados apontam uma mem?ria diferencial para os SE identificados como n?o confi?veis. A priori poder?amos constatar e aceitar isso como evid?ncia para a exist?ncia de um m?dulo espec?fico para a detec??o de trapa?a, uma vez que se lembrar de indiv?duos n?o confi?veis, de fato, ? fundamental para amenizar ou evitar os custos das intera??es com os mesmos no futuro. Entretanto, as an?lises subseq?entes, incluindo vari?veis inerentes a qualquer indiv?duo como o sexo e a tonalidade da cor da pele dos SE mostram a influ?ncia da sele??o sexual e dos fatores culturais na lembran?a de faces. Por exemplo, os SE do sexo eminino s?o mais bem lembradas pelos sujeitos experimentais do mesmo sexo e os SE confi?veis e com cor de pele mais escura s?o mais bem lembrados tanto pelos homens como mulheres participantes. Al?m disso, observamos que existe uma tend?ncia para lembran?a de eventos ou caracter?sticas percebidas como raras ou menos salientes na popula??o. Acreditamos que prestar aten??o ?quilo que ? diferente confere uma vantagem adaptativa, pois se a maioria da popula??o age de uma determinada forma ? f?cil ter uma estrat?gia de resposta ?s estrat?gias dos outros. Por?m, quando o padr?o ? raro ou err?tico ? preciso mudar o comportamento em resposta a esses eventos que saem do padr?o. Assim sendo, prestar aten??o ?quilo que ? diferente pode ter permitido os indiv?duos a usarem estrat?gias condicionais de maneira mais eficiente

Submitted by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2018-04-09T13:57:30Z No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Reprodução Não Autorizada.pdf: 47716 bytes, checksum: 0353d988c60b584cfc9978721c498a11 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2018-04-09T13:57:42Z (GMT) No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Reprodução Não Autorizada.pdf: 47716 bytes, checksum: 0353d988c60b584cfc9978721c498a11 (MD5) / Made available in DSpace on 2018-04-09T13:57:42Z (GMT). No. of bitstreams: 2 license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Reprodução Não Autorizada.pdf: 47716 bytes, checksum: 0353d988c60b584cfc9978721c498a11 (MD5) Previous issue date: 2012-09-24 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / This dissertation work presents the development of a new evaluation method for Haar classifiers designed to locate faces in images, in order to help researchers choose the best classifier for their own need, as well as to propose a new metric to evaluate future classifiers designed for face location. Face location, the primary step of the vision-based automated systems, finds the face area in the input image. An accurate location of the face is still a challenging task, but it enables a more efficient segmentation, making the identification and recognition of individuals more accurate. Viola-Jones framework has been widely used by researchers in order to detect the location of faces and objects in a given image. Face detection classifiers, used by Viola-Jones framework, are shared by the scientific and academic community. Nevertheless, few works have studied their accuracy. By applying 10 of these classifiers in two different face databases (FEI and Yale), an analysis was made and a new heuristic was created considering scores given to 22 facial landmarks. The results appear to be more accurate than the ones obtained by the only heuristic proposed by another work, developed so far. / Este trabalho de dissertação apresenta o desenvolvimento de um novo método de avaliação de classificadores Haar para localização de faces em imagens, com a finalidade de auxiliar pesquisadores na escolha do melhor classificador dentre os disponíveis atualmente, assim como oferecer uma nova métrica para avaliar futuros classificadores de localização facial. A localização facial, primeira etapa de sistemas biométricos pela face, limita as regiões das faces na imagem de entrada. A localização precisa da face ainda é uma tarefa desafiadora, e possibilita uma segmentação da área da face mais eficaz, aumentando a acurácia no processo de identificação e reconhecimento de indivíduos. O framework Viola-Jones tem sido amplamente utilizado por pesquisadores para detecção e localização de objetos e faces. Classificadores de detecção facial que utilizam o método Viola-Jones são compartilhados pela comunidade científica e acadêmica. Porém, pouco é discutido sobre a precisão de tais classificadores. Com a aplicação de 10 classificadores em dois bancos de faces distintas entre si (FEI e Yale), uma análise foi realizada e um novo método de avaliação foi proposto. Neste método uma nova métrica foi desenvolvida, levando em consideração scores dados a 22 pontos faciais. Os resultados obtidos mostraram-se mais precisos que a única metodologia de avaliação de classificadores de face Haar presente até o momento.