Seleção de features guiada por atenção visual em imagens com fóvea

Visual attention is a very important task in autonomous robotics, but, because of its complexity, the processing time required is significant. We propose an architecture for feature selection using foveated images that is guided by visual attention tasks and that reduces the processing time required to perform these tasks. Our system can be applied in bottom-up or top-down visual attention. The foveated model determines which scales are to be used on the feature extraction algorithm. The system is able to discard features that are not extremely necessary for the tasks, thus, reducing the processing time. If the fovea is correctly placed, then it is possible to reduce the processing time without compromising the quality of the tasks outputs. The distance of the fovea from the object is also analyzed. If the visual system loses the tracking in top-down attention, basic strategies of fovea placement can be applied. Experiments have shown that it is possible to reduce up to 60% the processing time with this approach. To validate the method, we tested it with the feature algorithm known as Speeded Up Robust Features (SURF), one of the most efficient approaches for feature extraction. With the proposed architecture, we can accomplish real time requirements of robotics vision, mainly to be applied in autonomous robotics

Influência do gênero na demanda visual por mapas em escalas diferentes em sistema de navegação e guia de rota

This work investigates the gender effect on visual demand of drivers for dynamic maps at different cartographic scales presented In-Vehicle Route Guidance and Navigation System (RGNS). A group of 52 subjects (26 males and 26 females) took part in an experiment performed in a low-cost driving simulator. the driver's task consisted of navigating in an unknown route using a RGNS prototype which presents maps at two different cartographic scales. This paper replicates the known phenomenon of significant relationships between gender and performance at visual-spatial tasks issue. Our results show that drivers of different genders present distinct levels of visual demand both due to the cartographic scales and maneuver complexity variation. These discussed results are based upon individual differences in terms of spatial ability and spatial anxiety.

Effetto crowding e dislessia evolutiva: un confronto inter e intra-linguistico

Crowding is defined as the negative effect obtained by adding visual distractors around a central target which has to be identified. Some studies have suggested the presence of a marked crowding effect in developmental dyslexia (e.g. Atkinson, 1991; Spinelli et al., 2002). Inspired by Spinelli’s (2002) experimental design, we explored the hypothesis that the crowding effect may affect dyslexics’ response times (RTs) and accuracy in identification tasks dealing with words, pseudowords, illegal non-words and symbolstrings. Moreover, our study aimed to clarify the relationship between the crowding phenomenon and the word-reading process, in an inter-language comparison perspective. For this purpose we studied twenty-two French dyslexics and twenty-two Italian dyslexics (total forty-four dyslexics), compared to forty-four subjects matched for reading level (22 French and 22 Italians) and forty-four chronological age-matched subjects (22 French and 22 Italians). Children were all tested on reading and cognitive abilities. Results showed no differences between French and Italian participants suggesting that performances were homogenous. Dyslexic children were all significantly impaired in words and pseudowords reading compared to their normal reading controls. Regarding the identification task with which we assessed crowding effect, both accuracy and RTs showed a lexicality effect which meant that the recognition of words was more accurate and faster in words than pseudowords, non-words and symbolstrings. Moreover, compared to normal readers, dyslexics’ RTs and accuracy were impaired only for verbal materials but not for non-verbal material; these results are in line with the phonological hypothesis (Griffiths & Snowling, 2002; Snowling, 2000; 2006) . RTs revealed a general crowding effect (RTs in the crowding condition were slower than those recorded in the isolated condition) affecting all the subjects’ performances. This effect, however, emerged to be not specific for dyslexics. Data didn’t reveal a significant effect of language, allowing the generalization of the obtained results. We also analyzed the performance of two subgroups of dyslexics, categorized according to their reading abilities. The two subgroups produced different results regarding the crowding effect and type of material, suggesting that it is meaningful to take into account also the heterogeneity of the dyslexia disorder. Finally, we also analyzed the relationship of the identification task with both reading and cognitive abilities. In conclusion, this study points out the importance of comparing visual tasks performances of dyslexic participants with those of their reading level-matched controls. This approach may improve our comprehension of the potential causal link between crowding and reading (Goswami, 2003).

A recuperação da informação da memória de trabalho: evidências baseadas em tarefas de imagem mental e de memória visual

A imagem mental e a memória visual têm sido consideradas como componentes distintos na codificação da informação, e associados a processos diferentes da memória de trabalho. Evidências experimentais mostram, por exemplo, que o desempenho em tarefas de memória baseadas na geração de imagem mentais (imaginação visual) sofre a interferência do ruído visual dinâmico (RVD), mas não se observa o mesmo efeito em tarefas de memória visual baseadas na percepção visual (memória visual). Embora várias evidências mostrem que tarefas de imaginação e de memória visual sejam baseadas em processos cognitivos diferentes, isso não descarta a possibilidade de utilizarem também processos em comum e que alguns resultados experimentais que apontam diferenças entre as duas tarefas resultem de diferenças metodológicas entre os paradigmas utilizados para estuda-las. Nosso objetivo foi equiparar as tarefas de imagem mental visual e memória visual por meio de tarefas de reconhecimento, com o paradigma de dicas retroativas espaciais. Sequências de letras romanas na forma visual (tarefa de memória visual) e acústicas (tarefa de imagem mental visual) foram apresentadas em quatro localizações espaciais diferentes. No primeiro e segundo experimento analisou-se o tempo do curso de recuperação tanto para o processo de imagem quanto para o processo de memória. No terceiro experimento, comparou-se a estrutura das representações dos dois componentes, por meio da apresentação do RVD durante a etapa de geração e recuperação. Nossos resultados mostram que não há diferenças no armazenamento da informação visual durante o período proposto, porém o RVD afeta a eficiência do processo de recuperação, isto é o tempo de resposta, sendo a representação da imagem mental visual mais suscetível ao ruído. No entanto, o processo temporal da recuperação é diferente para os dois componentes, principalmente para imaginação que requer mais tempo para recuperar a informação do que a memória. Os dados corroboram a relevância do paradigma de dicas retroativas que indica que a atenção espacial é requisitada em representações de organização espacial, independente se são visualizadas ou imaginadas.

Visual signs and symptoms of dementia with Lewy bodies

Dementia with Lewy bodies ('Lewy body dementia' or 'diffuse Lewy body disease') (DLB) is the second most common form of dementia to affect elderly people, after Alzheimer's disease. A combination of the clinical symptoms of Alzheimer's disease and Parkinson's disease is present in DLB and the disorder is classified as a 'parkinsonian syndrome', a group of diseases which also includes Parkinson's disease, progressive supranuclear palsy, corticobasal degeneration and multiple system atrophy. Characteristics of DLB are fluctuating cognitive ability with pronounced variations in attention and alertness, recurrent visual hallucinations and spontaneous motor features, including akinesia, rigidity and tremor. In addition, DLB patients may exhibit visual signs and symptoms, including defects in eye movement, pupillary function and complex visual functions. Visual symptoms may aid the differential diagnoses of parkinsonian syndromes. Hence, the presence of visual hallucinations supports a diagnosis of Parkinson's disease or DLB rather than progressive supranuclear palsy. DLB and Parkinson's disease may exhibit similar impairments on a variety of saccadic and visual perception tasks (visual discrimination, space-motion and object-form recognition). Nevertheless, deficits in orientation, trail-making and reading the names of colours are often significantly greater in DLB than in Parkinson's disease. As primary eye-care practitioners, optometrists should be able to work with patients with DLB and their carers to manage their visual welfare.

A visual system for distributed mosaics using an auvs fleet

The use of teams of Autonomous Underwater Vehicles for visual inspection tasks is a promising robotic field. The images captured by different robots can be also to aid in the localization/navigation of the fleet. In a previous work, a distributed localization system was presented based on the use of Augmented States Kalman Filter through the visual maps obtained by the fleet. In this context, this paper details a system for on-line construction of visual maps and its use to aid the localization and navigation of the robots. Different aspects related to the capture, treatment and construction of mosaics by fleets of robots are presented. The developed system can be executed on-line on different robotic platforms. The paper is concluded with a series of tests and analyses aiming at to system validation.

V1-inspired features induce a weighted margin in SVMs

Image representations derived from simplified models of the primary visual cortex (V1), such as HOG and SIFT, elicit good performance in a myriad of visual classification tasks including object recognition/detection, pedestrian detection and facial expression classification. A central question in the vision, learning and neuroscience communities regards why these architectures perform so well. In this paper, we offer a unique perspective to this question by subsuming the role of V1-inspired features directly within a linear support vector machine (SVM). We demonstrate that a specific class of such features in conjunction with a linear SVM can be reinterpreted as inducing a weighted margin on the Kronecker basis expansion of an image. This new viewpoint on the role of V1-inspired features allows us to answer fundamental questions on the uniqueness and redundancies of these features, and offer substantial improvements in terms of computational and storage efficiency.

A sparse hybrid map for vision-guided mobile robots

This paper introduces a minimalistic approach to produce a visual hybrid map of a mobile robot’s working environment. The proposed system uses omnidirectional images along with odometry information to build an initial dense posegraph map. Then a two level hybrid map is extracted from the dense graph. The hybrid map consists of global and local levels. The global level contains a sparse topological map extracted from the initial graph using a dual clustering approach. The local level contains a spherical view stored at each node of the global level. The spherical views provide both an appearance signature for the nodes, which the robot uses to localize itself in the environment, and heading information when the robot uses the map for visual navigation. In order to show the usefulness of the map, an experiment was conducted where the map was used for multiple visual navigation tasks inside an office workplace.

Place categorization and semantic mapping on a mobile robot

In this paper we focus on the challenging problem of place categorization and semantic mapping on a robot with-out environment-specific training. Motivated by their ongoing success in various visual recognition tasks, we build our system upon a state-of-the-art convolutional network. We overcome its closed-set limitations by complementing the network with a series of one-vs-all classifiers that can learn to recognize new semantic classes online. Prior domain knowledge is incorporated by embedding the classification system into a Bayesian filter framework that also ensures temporal coherence. We evaluate the classification accuracy of the system on a robot that maps a variety of places on our campus in real-time. We show how semantic information can boost robotic object detection performance and how the semantic map can be used to modulate the robot’s behaviour during navigation tasks. The system is made available to the community as a ROS module.

Deep learning review and its applications

Deep neural networks have recently gained popularity for improv- ing state-of-the-art machine learning algorithms in diverse areas such as speech recognition, computer vision and bioinformatics. Convolutional networks especially have shown prowess in visual recognition tasks such as object recognition and detection in which this work is focused on. Mod- ern award-winning architectures have systematically surpassed previous attempts at tackling computer vision problems and keep winning most current competitions. After a brief study of deep learning architectures and readily available frameworks and libraries, the LeNet handwriting digit recognition network study case is developed, and lastly a deep learn- ing network for playing simple videogames is reviewed.

Aproximação fisionômica pericial através de função de base radial hermitiana

A aproximação fisionômica é o método que busca, a partir do crânio, simular a fotografia de um indivíduo quando em vida. Deve ser empregada como último recurso, na busca de desaparecidos, quando não houver possibilidade de aplicação de um método válido de identificação. O objetivo deste estudo foi obter a aproximação fisionômica, a partir de um crânio seco e de tomografia computadorizada multislice de indivíduos vivos, através da função de base radial hermitiana (FBRH). Constituiu-se também em avaliar o resultado da mesma quanto ao reconhecimento. Na primeira etapa do estudo, foi utilizada a imagem escaneada de um crânio seco, de origem desconhecida, com o intuito de avaliar se a quantidade de pontos obtidos seria suficiente para aplicação da FBRH e consequente reconstrução da superfície facial. Na segunda fase, foram utilizadas três tomografias de indivíduos vivos, para análise da semelhança alcançada entre a face escaneada e as aproximações faciais. Nesta etapa, foi aplicada uma associação de diferentes metodologias já publicadas, para reconstrução de uma mesma região da face, a partir de um mesmo crânio. Na última etapa, foram simuladas situações de reconhecimento com familiares e amigos dos indivíduos doadores das tomografias. Observou-se que a metodologia de FBRH pode ser empregada em aproximação fisionômica. Houve reconhecimento positivo nos três sujeitos estudados, sendo que, em dois deles, os resultados foram ainda mais significativos. Desta forma, conclui-se que a metodologia é rápida, objetiva e proporciona o reconhecimento. Esta permite a criação de múltiplas versões de aproximações fisionômicas a partir do mesmo crânio, o que amplia as possibilidades de reconhecimento. Observou-se ainda que a técnica não exige habilidade artística do profissional.

建立猕猴Go/NoGo视觉分辨任务的方法

在认知神经科学研究中,Go/NoGo模型是一种非常有效的研究方法。在本试验中,以两只猕猴为研究对象,采用Go/NoGo模型,以不同的视觉线索作为刺激来研究相关认知行为。结果表明猕猴能够很快学会Go/NoGo视觉分辨任务,而且对NoGo任务的完成要优于对Go任务的完成。本实验建立了一种有效的猕猴Go/NoGo视觉分辨实验的方法及计算机控制系统,为进一步记录神经元活动建立了良好的基础。

Simultaneous colour search renders other object properties less salient

A common approach to visualise multidimensional data sets is to map every data dimension to a separate visual feature. It is generally assumed that such visual features can be judged independently from each other. However, we have recently shown that interactions between features do exist [Hannus et al. 2004; van den Berg et al. 2005]. In those studies, we first determined individual colour and size contrast or colour and orientation contrast necessary to achieve a fixed level of discrimination performance in single feature search tasks. These contrasts were then used in a conjunction search task in which the target was defined by a combination of a colour and a size or a colour and an orientation. We found that in conjunction search, despite the matched feature discriminability, subjects significantly more often chose an item with the correct colour than one with correct size or orientation. This finding may have consequences for visualisation: the saliency of information coded by objects' size or orientation may change when there is a need to simultaneously search for colour that codes another aspect of the information. In the present experiment, we studied whether a colour bias can also be found in a more complex and continuous task, Subjects had to search for a target in a node-link diagram consisting of SO nodes, while their eye movements were being tracked, Each node was assigned a random colour and size (from a range of 10 possible values with fixed perceptual distances). We found that when we base the distances on the mean threshold contrasts that were determined in our previous experiments, the fixated nodes tend to resemble the target colour more than the target size (Figure 1a). This indicates that despite the perceptual matching, colour is judged with greater precision than size during conjunction search. We also found that when we double the size contrast (i.e. the distances between the 10 possible node sizes), this effect disappears (Figure 1b). Our findings confirm that the previously found decrease in salience of other features during colour conjunction search is also present in more complex (more 'visualisation- realistic') visual search tasks. The asymmetry in visual search behaviour can be compensated for by manipulating step sizes (perceptual distances) within feature dimensions. Our results therefore also imply that feature hierarchies are not completely fixed and may be adapted to the requirements of a particular visualisation. Copyright © 2005 by the Association for Computing Machinery, Inc.

听力障碍与听力正常群体视觉注意发展及其机制的比较

One of the most important functions in the individual development is the interaction and integration of each sensory input. There exist two competing theories, i.e. the deficiency theory and the compensatory theory, regarding the origin and nature of changes in visual functions observed after auditory deprivation. The deficiency theory proposed that integrative processes are essential for normal development. In contrast, the compensatory theory stated that the loss of one sense may be met by a greater reliance upon, therefore an enhancement of the remaining senses. Given that hearing impaired children’s learning depends primarily on visual information, it is important to recognize the differences of visual attention between them and their hearing age-mates. Differences among age groups could exist in either selectivity or sustained attention. Study 1 and study 2 explored the selective and sustained attention development of hearing impaired and hearing students with average cognitive ability, aged from 7 years to college students. The analysis and discussion of the results are based on the visual attention development as well as deficiency theory and compensatory theory. According to the results of the study 1 and study 2, the spatial distribution and controlling of the visual attention between hearing impaired and hearing students were also investigated in the study 3 and study 4. The present work showed that: Firstly, both hearing impaired and hearing participants had the similar developmental trajectory of the sustained attention. The ability of children’s sustained attention appeared to improve with age, and in adolescence it reached the peak. The hearing impaired participants had the comparable sustained attention skills to the matched hearing ones. Besides, the results of the hearing impaired participants showed that they could maintain their attention and vigilance on the current task over the observation period. Secondly, group differences of visual attention development were found between hearing impaired and hearing participants. In the childhood, the visual attention developmental speed of the hearing impaired children was slower than that of the hearing ones. The selective attention skill of the hearing impaired were not comparable to the hearing ones, however, their selective skill improved with age, so in the adulthood, hearing impaired students showed the slight advantage in the selective attention skill over the hearing ones. Thirdly, hearing impaired and hearing participants showed the similar spatial distribution in the attention resources. In the low perceptual load condition, both participants were suffered great interference of the distrator at the fixation. In contrast, in the high perceptual load condition, hearing impaired adults were suffered more interference of the peripheral distractor, which suggested that they distributed more attention resources to the peripheral field when faced difficult tasks. Fourthly, both groups showed similar processing in the visual attention tasks. That is, they both searched the target with only the color feature in a parallel way, but in a serial way while processing orientation feature and the features with the combination of the color and orientation. Furthermore, the results indicated that two groups show similar ways in the attention controlling. In summary, the present study showed that visual attention development was dependent upon the integration of multimodal sensory information. Because of the interaction and integration of the input from various sensory, it has a negative impact on the intact sensory at the early stage of one sensory loss, however, it can better the functions of other intact sensory gradually with development and practice.

非言语学习不良青少年空间关系推理及其影响因素研究

It has reported that individuals with nonverbal learning disabilities (NLD) have deficits in visual-spatial organization and strengths in rote language abilities. At present, there are few studies on higher order cognitive abilities of adolescents with NLD, such as the reasoning about spatial relations. The study sampled three groups: a normal group (a control group, C), a nonverbal learning disabilities group (NLD), and a verbal learning disabilities group (VLD). The aim of this study was to examine spatial and nonspatial relation reasoning abilities in adolescents with NLD under figure and word conditions, and assessed the relative involvement of different working memory components in four types of reasoning tasks: reasoning about figure-spatial, figure-nonspatial, verbal-spatial, and verbal-nonspatial relations. Using the double-tasks methodology, visual, spatial, central-executive, and phonological loads were realized. We tried to find how working memory components impact on adolescents with NLD spatial and nonspatial reasoning. The main results of present research are as follows. (1) The NLD group didn’t differ from normal group on reasoning about figure-nonspatial relations. The NLD group scored lower than the C group in spatial problems. So, adolescents with NLD showed a dissociation between spatial and non-spatial relation reasoning. They scored higher in non-spatial problems than in spatial ones. Adolescents with VLD developed well in reasoning about figure-nonspatial relations, but showed deficits in other three tasks. (2) For each reasoning task, the difficult of four types of reasoning problem had different changing trend. For figure and verbal spatial problems, mental model approach can interpret performance of the four problems well. For verbal nonspatial problems, a logical rule approach can interpret performance of the four problems well. (3) Adolescents with NLD did not differ from adolescents with VLD and normal adolescents in phonological, central-executive, and visual dual tasks. But the NLD group had lower performance than the other two groups in spatial dual task. The results showed a dissociation between visual and spatial working memory in NLD group. The VLD group only experienced deficits in central-executive subsystem. (4) The studies found that spatial reasoning mainly loaded spatial working memory, whist the involvement of spatial resources in nonspatial reasoning was little. Visual working memory mainly involved in reasoning about spatial and figure-nonspatial relations, especially in figure-nonspatial problems, and had few impacts on verbal-nonspatial reasoning. Central executive system was involved in all reasoning tasks. The role of phonological loop in the reasoning tasks required further explored. (5) According to the findings, we concluded that the deficits in spatial working memory resulted in poor spatial reasoning abilities for teenagers with NLD, whist because of the limited central executive capability, teenagers with VLD showed poor reasoning abilities. (6) The three groups can used multiple strategies during the reasoning process. They didn’t differ from each other in reasoning strategies. They all used mental model strategy to solve figure and verbal spatial problems, and used logic rule strategy to solve verbal nonspatial problems.