View-Based Strategies for 3D Object Recognition

A persistent issue of debate in the area of 3D object recognition concerns the nature of the experientially acquired object models in the primate visual system. One prominent proposal in this regard has expounded the use of object centered models, such as representations of the objects' 3D structures in a coordinate frame independent of the viewing parameters [Marr and Nishihara, 1978]. In contrast to this is another proposal which suggests that the viewing parameters encountered during the learning phase might be inextricably linked to subsequent performance on a recognition task [Tarr and Pinker, 1989; Poggio and Edelman, 1990]. The 'object model', according to this idea, is simply a collection of the sample views encountered during training. Given that object centered recognition strategies have the attractive feature of leading to viewpoint independence, they have garnered much of the research effort in the field of computational vision. Furthermore, since human recognition performance seems remarkably robust in the face of imaging variations [Ellis et al., 1989], it has often been implicitly assumed that the visual system employs an object centered strategy. In the present study we examine this assumption more closely. Our experimental results with a class of novel 3D structures strongly suggest the use of a view-based strategy by the human visual system even when it has the opportunity of constructing and using object-centered models. In fact, for our chosen class of objects, the results seem to support a stronger claim: 3D object recognition is 2D view-based.

Face processing in humans is compatible with a simple shape-based model of vision

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.

Gender differences among undergraduates in the recognition of emotional facial expressions

The perceptive accuracy of university students was compared between men and women, from sciences and humanities courses, to recognize emotional facial expressions. emotional expressions have had increased interest in several areas involved with human interaction, reflecting the importance of perceptive skills in human expression of emotions for the effectiveness of communication. Two tests were taken: one was a quick exposure (0.5 s) of 12 faces with an emotional expression, followed by a neutral face. subjects had to tell if happiness, sadness, anger, fear, disgust or surprise was flashed, and each emotion was shown twice, at random. on the second test 15 faces with the combination of two emotional expressions were shown without a time limit, and the subject had to name one of the emotions of the previous list. in this study, women perceived sad expressions better while men realized more happy faces. there was no significant difference in other emotions detection like anger, fear, surprise, disgust. Students of humanities and sciences areas of both sexes, when compared, had similar capacities to perceive emotional expressions

Emotion recognition in multiple sclerosis

The effect of multiple sclerosis (MS) on the ability to identify emotional expressions in faces was investigated, and possible associations with patients’ characteristics were explored. 56 non-demented MS patients and 56 healthy subjects (HS) with similar demographic characteristics performed an emotion recognition task (ERT), the Benton Facial Recognition Test (BFRT), and answered the Hospital Anxiety and Depression Scale (HADS). Additionally, MS patients underwent a neurological examination and a comprehensive neuropsychological evaluation. The ERT consisted of 42 pictures of faces (depicting anger, disgust, fear, happiness, sadness, surprise and neutral expressions) from the NimStim set. An iViewX high-speed eye tracker was used to record eye movements during ERT. The fixation times were calculated for two regions of interest (i.e., eyes and rest of the face). No significant differences were found between MS and HC on ERT’s behavioral and oculomotor measures. Bivariate and multiple regression analyses revealed significant associations between ERT’s behavioral performance and demographic, clinical, psychopathological, and cognitive measures.

A Reabilitação face à Sustentabilidade : Casa Burguesa do Porto

Os centros históricos são o resultado de um parque habitacional corrente, com uma grande uniformidade, coerência construtiva e espaços urbanos de modo a proporcionar uma identidade própria. Um dos grandes problemas urbanos são as vastas áreas degradadas do ponto de vista arquitetónico como também social, cultural e económico. Conhecidos os principais problemas dos edifícios de habitação e suas causas, para sua resolução existe necessidade de adaptar o processo de reabilitação tradicional ao conceito de sustentabilidade. Estes dois assuntos conjugados são atualmente emergentes, devido à necessidade de reabilitação do parque habitacional, nomeadamente centros históricos. A casa burguesa do Porto apresenta um elevado grau de degradação, porém apesar de existentes as ações de intervenção que sobre ela reincidem, não são proporcionais à necessidade atual. A cidade do Porto é marcada, maioritariamente, pelas Casas Burguesas, das quais foram executadas segundo padrões de conforto e de utilização da época. É, então, fundamental um estudo pormenorizado da casa burguesa do Porto, avaliando o contexto em que esta se encere (na cidade e na respetiva habitação), os subsistemas construtivos e as praticas de reabilitação.

Host plant recognition by the root feeding clover weevil, Sitona lepidus (Coleoptera : Curculionidae)

This study investigated the ability of neonatal larvae of the root-feeding weevil, Sitona lepidus Gyllenhal, to locate white clover Trifolium repens L. (Fabaceae) roots growing in soil and to distinguish them from the roots of other species of clover and a co-occurring grass species. Choice experiments used a combination of invasive techniques and the novel technique of high resolution X-ray microtomography to non-invasively track larval movement in the soil towards plant roots. Burrowing distances towards roots of different plant species were also examined. Newly hatched S. lepidus recognized T. repens roots and moved preferentially towards them when given a choice of roots of subterranean clover, Trifolium subterraneum L. (Fabaceae), strawberry clover Trifolium fragiferum L. (Fabaceae), or perennial ryegrass Lolium perenne L. (Poaceae). Larvae recognized T. repens roots, whether released in groups of five or singly, when released 25 mm (meso-scale recognition) or 60 mm (macro-scale recognition) away from plant roots. There was no statistically significant difference in movement rates of larvae.

Asymmetric synthesis of (1R,2S,3R)-3-methylcispentacin and (1S,2S,3R)-3-methyltranspentacin by kinetic resolution of tert-butyl (+/-)-3-methylcyclopentene-1-carboxylate

Conjugate addition of lithium dibenzylamide to tert-butyl (+/-)-3-methylcyclopentene-1-carboxylate occurs with high levels of stereocontrol, with preferential addition of lithium dibenzylamide to the face of the cyclic alpha,beta-unsaturated acceptor anti- to the 3-methyl substituent. High levels of enantiorecognition are observed between tert-butyl (+/-)-3-methylcyclopentene-1-carboxylate and an excess of lithium (+/-)-N-benzyl-N-alpha-methylbenzylamide (10 eq.) (E > 140) in their mutual kinetic resolution, while the kinetic resolution of tert-butyl (+/-)-3-methylcyclopentene-1-carboxylate with lithium (S)-N-benzyl-N-alpha-methylbenzylamide proceeds to give, at 51% conversion, tert-butyl (1R, 2S, 3R,alphaS)-3-methyl-2-N-benzyl-N-alpha-methylbenzylaminocyclopentane-1-c arboxylate consistent with E > 130, and in 39% yield and 99 +/- 0.5% de after purification. Subsequent deprotection by hydrogenolysis and ester hydrolysis gives (1R, 2S, 3R)-3-methylcispentacin in > 98% de and 98 +/- 1% ee. Selective epimerisation of tert-butyl (1R, 2S, 3R, alphaS)-3-methyl-2-N- benzyl-N-alpha-methylbenzylaminocyclopentane-1-carboxylate by treatment with (KOBu)-Bu-t in (BuOH)-Bu-t gives tert-butyl (1S, 2S, 3R, alphaS)-3-methyl-2-N-benzyl-N-alpha-methylbenzylaminocyclopentane-1-carb oxylate in quantitative yield and in > 98% de, with subsequent deprotection by hydrogenolysis and ester hydrolysis giving (1S, 2S, 3R)-3-methyltranspentacin hydrochloride in > 98% de and 97 +/- 1% ee.

Face verification using Gabor wavelets and AdaBoost

This paper presents a new face verification algorithm based on Gabor wavelets and AdaBoost. In the algorithm, faces are represented by Gabor wavelet features generated by Gabor wavelet transform. Gabor wavelets with 5 scales and 8 orientations are chosen to form a family of Gabor wavelets. By convolving face images with these 40 Gabor wavelets, the original images are transformed into magnitude response images of Gabor wavelet features. The AdaBoost algorithm selects a small set of significant features from the pool of the Gabor wavelet features. Each feature is the basis for a weak classifier which is trained with face images taken from the XM2VTS database. The feature with the lowest classification error is selected in each iteration of the AdaBoost operation. We also address issues regarding computational costs in feature selection with AdaBoost. A support vector machine (SVM) is trained with examples of 20 features, and the results have shown a low false positive rate and a low classification error rate in face verification.

Towards semantic resolution of security in ambient environments

Driven by new network and middleware technologies such as mobile broadband, near-field communication, and context awareness the so-called ambient lifestyle will foster innovative use cases in different domains. In the EU project Hydra high-level security, trust and privacy concerns such as loss of control, profiling and surveillance are considered at the outset. At the end of this project the. Hydra middleware development platform will have been designed so as to enable developers to realise secure ambient scenarios. This paper gives a short introduction to the Hydra project and its approach to ensure security by design. Based on the results of a focus group analysis of the user domain "building automation" typical threats are evaluated and their risks are assessed. Then, specific security requirements with respect to security, privacy, and trust are derived in order to incorporate them into the Hydra Security Meta-Model. How concepts such as context, semantic resolution of security, and virtualisation support the overall Hydra approach will be introduced and illustrated on the basis of it technical building automation scenario.

Enhancing change detection in low-quality surveillance footage using markov random fields

Urban surveillance footage can be of poor quality, partly due to the low quality of the camera and partly due to harsh lighting and heavily reflective scenes. For some computer surveillance tasks very simple change detection is adequate, but sometimes a more detailed change detection mask is desirable, eg, for accurately tracking identity when faced with multiple interacting individuals and in pose-based behaviour recognition. We present a novel technique for enhancing a low-quality change detection into a better segmentation using an image combing estimator in an MRF based model.

Robust real-time tracking for visual surveillance

This paper describes a real-time multi-camera surveillance system that can be applied to a range of application domains. This integrated system is designed to observe crowded scenes and has mechanisms to improve tracking of objects that are in close proximity. The four component modules described in this paper are (i) motion detection using a layered background model, (ii) object tracking based on local appearance, (iii) hierarchical object recognition, and (iv) fused multisensor object tracking using multiple features and geometric constraints. This integrated approach to complex scene tracking is validated against a number of representative real-world scenarios to show that robust, real-time analysis can be performed. Copyright (C) 2007 Hindawi Publishing Corporation. All rights reserved.

Communication mechanisms and middleware for distributed video surveillance

A new generation of advanced surveillance systems is being conceived as a collection of multi-sensor components such as video, audio and mobile robots interacting in a cooperating manner to enhance situation awareness capabilities to assist surveillance personnel. The prominent issues that these systems face are: the improvement of existing intelligent video surveillance systems, the inclusion of wireless networks, the use of low power sensors, the design architecture, the communication between different components, the fusion of data emerging from different type of sensors, the location of personnel (providers and consumers) and the scalability of the system. This paper focuses on the aspects pertaining to real-time distributed architecture and scalability. For example, to meet real-time requirements, these systems need to process data streams in concurrent environments, designed by taking into account scheduling and synchronisation. The paper proposes a framework for the design of visual surveillance systems based on components derived from the principles of Real Time Networks/Data Oriented Requirements Implementation Scheme (RTN/DORIS). It also proposes the implementation of these components using the well-known middleware technology Common Object Request Broker Architecture (CORBA). Results using this architecture for video surveillance are presented through an implemented prototype.

Assembling novel protein folds from super-secondary structural fragments

The results of applying a fragment-based protein tertiary structure prediction method to the prediction of 14 CASP5 target domains are described. The method is based on the assembly of supersecondary structural fragments taken from highly resolved protein structures using a simulated annealing algorithm. A number of good predictions for proteins with novel folds were produced, although not always as the first model. For two fold recognition targets, FRAGFOLD produced the most accurate model in both cases, despite the fact that the predictions were not based on a template structure. Although clear progress has been made in improving FRAGFOLD since CASP4, the ranking of final models still seems to be the main problem that needs to be addressed before the next CASP experiment

Robust abandoned object detection integrating wide area visual surveillance and social context

This paper presents a video surveillance framework that robustly and efficiently detects abandoned objects in surveillance scenes. The framework is based on a novel threat assessment algorithm which combines the concept of ownership with automatic understanding of social relations in order to infer abandonment of objects. Implementation is achieved through development of a logic-based inference engine based on Prolog. Threat detection performance is conducted by testing against a range of datasets describing realistic situations and demonstrates a reduction in the number of false alarms generated. The proposed system represents the approach employed in the EU SUBITO project (Surveillance of Unattended Baggage and the Identification and Tracking of the Owner).

A high resolution late Holocene palaeo environmental record from the central Adriatic Sea

A multi-proxy study of a Holocene sediment core (RF 93-30) from the western flank of the central Adriatic, in 77 m of water, reveals a sequence of changes in terrestrial vegetation, terrigenous sediment input and benthic fauna, as well as evidence for variations in sea surface temperature spanning most of the last 7000 yr. The chronology of sedimentation is based on several lines of evidence, including AMS 14C dates of foraminifera extracted from the core, palaeomagnetic secular variation, pollen indicators and dated tephra. The temporal resolution increases towards the surface and, for some of the properties measured, is sub-decadal for the last few centuries. The main changes recorded in vegetation, sedimentation and benthic foraminiferal assemblages appear to be directly related to human activity in the sediment source area, which includes the Po valley and the eastern flanks of the central and northern Appenines. The most striking episodes of deforestation and expanding human impact begin around 3600 BP (Late Bronze Age) and 700 BP (Medieval) and each leads to an acceleration in mass sedimentation and an increase in the proportion of terrigenous material, reflecting the response of surface processes to widespread forest clearance and cultivation. Although human impact appears to be the proximal cause of these changes, climatic effects may also have been important. During these periods, signs of stress are detectable in the benthic foram morphotype assemblages. Between these two periods of increased terrigeneous sedimentation there is smaller peak in sedimentation rate around 2400BP which is not associated with evidence for deforestation, shifts in the balance between terrigenous and authigenic sedimentation, or changes in benthic foraminifera. The mineral magnetic record provides a sensitive indicator of changing sediment sources: during forested periods of reduced terrigenous input it is dominated by authigenic bacterial magnetite, whereas during periods of increased erosion, anti-ferromagetic minerals (haematite and/or goethite) become more important, as well as both paramagnetic minerals and super-paramagnetic magnetite. Analysis of the alkenone, U37k′, record provides an indication of possible changes in sea surface temperature during the period, but it is premature to place too much reliance on these inferred changes until the indirect effects of past changes in the depth of the halocline and in circulation have been more fully evaluated. The combination of methods used and the results obtained illustrate the potential value of such high resolution near-shore marine sedimentary sequences for recording wide-scale human impact, documenting the effects of this on marine sedimentation and fauna and, potentially, disentangling evidence for human activities from that for past changes in climate.