ReadingAct RGB-D action dataset and human action recognition from local features

For general home monitoring, a system should automatically interpret people’s actions. The system should be non-intrusive, and able to deal with a cluttered background, and loose clothes. An approach based on spatio-temporal local features and a Bag-of-Words (BoW) model is proposed for single-person action recognition from combined intensity and depth images. To restore the temporal structure lost in the traditional BoW method, a dynamic time alignment technique with temporal binning is applied in this work, which has not been previously implemented in the literature for human action recognition on depth imagery. A novel human action dataset with depth data has been created using two Microsoft Kinect sensors. The ReadingAct dataset contains 20 subjects and 19 actions for a total of 2340 videos. To investigate the effect of using depth images and the proposed method, testing was conducted on three depth datasets, and the proposed method was compared to traditional Bag-of-Words methods. Results showed that the proposed method improves recognition accuracy when adding depth to the conventional intensity data, and has advantages when dealing with long actions.

Microbial Recognition Via Toll-Like Receptor-Dependent and -Independent Pathways Determines the Cytokine Response of Murine Dendritic Cell Subsets to CD40 Triggering

Dendritic cells (DC) can produce Th-polarizing cytokines and direct the class of the adaptive immune response. Microbial stimuli, cytokines, chemokines, and T cell-derived signals all have been shown to trigger cytokine synthesis by DC, but it remains unclear whether these signals are functionally equivalent and whether they determine the nature of the cytokine produced or simply initiate a preprogrammed pattern of cytokine production, which may be DC subtype specific. Here, we demonstrate that microbial and T cell-derived stimuli can synergize to induce production of high levels of IL-12 p70 or IL-10 by individual murine DC subsets but that the choice of cytokine is dictated by the microbial pattern recognition receptor engaged. We show that bacterial components such as CpG-containing DNA or extracts from Mycobacterium tuberculosis predispose CD8alpha(+) and CD8alpha(-)CD4(-) DC to make IL-12 p70. In contrast, exposure of CD8alpha(+), CD4(+) and CD8alpha(-)CD4(-) DC to heat-killed yeasts leads to production of IL-10. In both cases, secretion of high levels of cytokine requires a second signal from T cells, which can be replaced by CD40 ligand. Consistent with their differential effects on cytokine production, extracts from M. tuberculosis promote IL-12 production primarily via Toll-like receptor 2 and an MyD88-dependent pathway, whereas heat-killed yeasts activate DC via a Toll-like receptor 2-, MyD88-, and Toll/IL-1R domain containing protein-independent pathway. These results show that T cell feedback amplifies innate signals for cytokine production by DC and suggest that pattern recognition rather than ontogeny determines the production of cytokines by individual DC subsets.

THE ROLE OF INNATE IMMUNITY IN SEPTIC ACUTE KIDNEY INJURIES

Acute kidney injury (AKI) is an important clinical syndrome characterized by abnormalities in the hydroelectrolytic balance. Because of high rates of morbidity and mortality (from 15% to 60%) associated with AKI, the study of its pathophysiology is critical in searching for clinical targets and therapeutic strategies. Severe sepsis is the major cause of AKI. The host response to sepsis involves an inflammatory response, whereby the pathogen is initially sensed by innate immune receptors (pattern recognition receptors [PRRs]). When it persists, this immune response leads to secretion of proinflammatory products that induce organ dysfunction such as renal failure and consequently increased mortality. Moreover, the injured tissue releases molecules resulting from extracellular matrix degradation or dying cells that function as alarmines, which are recognized by PRR in the absence of pathogens in a second wave of injury. Toll-like receptors (TLRs) and NOD-like receptors (NLRs) are the best characterized PRRs. They are expressed in many cell types and throughout the nephron. Their activation leads to translocation of nuclear factors and synthesis of proinflammatory cytokines and chemokines. TLRs` signaling primes the cells for a robust inflammatory response dependent on NLRs; the interaction of TLRs and NLRs gives rise to the multiprotein complex known as the inflammasome, which in turn activates secretion of mature interleukin 1 beta and interleukin 18. Experimental data show that innate immune receptors, the inflammasome components, and proinflammatory cytokines play crucial roles not only in sepsis, but also in organ-induced dysfunction, especially in the kidneys. In this review, we discuss the significance of the innate immune receptors in the development of acute renal injury secondary to sepsis.

A low-cost 3D reconstruction system using a single-shot projection of a pattern matrix

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

CENSUS HISTOGRAMS: A SIMPLE FEATURE EXTRACTION and MATCHING APPROACH FOR FACE RECOGNITION

Most face recognition approaches require a prior training where a given distribution of faces is assumed to further predict the identity of test faces. Such an approach may experience difficulty in identifying faces belonging to distributions different from the one provided during the training. A face recognition technique that performs well regardless of training is, therefore, interesting to consider as a basis of more sophisticated methods. In this work, the Census Transform is applied to describe the faces. Based on a scanning window which extracts local histograms of Census Features, we present a method that directly matches face samples. With this simple technique, 97.2% of the faces in the FERET fa/fb test were correctly recognized. Despite being an easy test set, we have found no other approaches in literature regarding straight comparisons of faces with such a performance. Also, a window for further improvement is presented. Among other techniques, we demonstrate how the use of SVMs over the Census Histogram representation can increase the recognition performance.

Robust and fast vowel recognition using optimum-path forest

The applications of Automatic Vowel Recognition (AVR), which is a sub-part of fundamental importance in most of the speech processing systems, vary from automatic interpretation of spoken language to biometrics. State-of-the-art systems for AVR are based on traditional machine learning models such as Artificial Neural Networks (ANNs) and Support Vector Machines (SVMs), however, such classifiers can not deal with efficiency and effectiveness at the same time, existing a gap to be explored when real-time processing is required. In this work, we present an algorithm for AVR based on the Optimum-Path Forest (OPF), which is an emergent pattern recognition technique recently introduced in literature. Adopting a supervised training procedure and using speech tags from two public datasets, we observed that OPF has outperformed ANNs, SVMs, plus other classifiers, in terms of training time and accuracy. ©2010 IEEE.

Improving face recognition with multispectral fusion and support vector machines

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Structure-activity relationship of compounds which are anti-schistosomiasis active

O método do orbital molecular AM1 foi empregado para calcular um conjunto de descritores moleculares para vinte neolignanas sintéticas com atividade anti-esquistossomose. O método de reconhecimento de padrão (análise de componentes principais ACP, análise de conglomerados AC e análise de discriminante) foi utilizado para obter a relação entre a estrutura molecular e a atividade biológica. O conjunto de moléculas foi classificado em dois grupos de acordo com seus graus de atividade biológica. Estes resultados permitem que, projete-se racionalmente novos compostos, potenciais candidatos à síntese e à avaliação biológica.

Rank Aggregation for Pattern Classifier Selection in Remote Sensing Images

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Multi-q pattern analysis: A case study in image classification

This paper compares the effectiveness of the Tsallis entropy over the classic Boltzmann-Gibbs-Shannon entropy for general pattern recognition, and proposes a multi-q approach to improve pattern analysis using entropy. A series of experiments were carried out for the problem of classifying image patterns. Given a dataset of 40 pattern classes, the goal of our image case study is to assess how well the different entropies can be used to determine the class of a newly given image sample. Our experiments show that the Tsallis entropy using the proposed multi-q approach has great advantages over the Boltzmann-Gibbs-Shannon entropy for pattern classification, boosting image recognition rates by a factor of 3. We discuss the reasons behind this success, shedding light on the usefulness of the Tsallis entropy and the multi-q approach. (C) 2012 Elsevier B.V. All rights reserved.

An evolutive approach for smile recognition in video sequences

Facial expression recognition is one of the most challenging research areas in the image recognition ¯eld and has been actively studied since the 70's. For instance, smile recognition has been studied due to the fact that it is considered an important facial expression in human communication, it is therefore likely useful for human–machine interaction. Moreover, if a smile can be detected and also its intensity estimated, it will raise the possibility of new applications in the future

Deep learning for computer vision: a comparison between convolutional neural networks and hierarchical temporal memories on object recognition tasks

In recent years, Deep Learning techniques have shown to perform well on a large variety of problems both in Computer Vision and Natural Language Processing, reaching and often surpassing the state of the art on many tasks. The rise of deep learning is also revolutionizing the entire field of Machine Learning and Pattern Recognition pushing forward the concepts of automatic feature extraction and unsupervised learning in general. However, despite the strong success both in science and business, deep learning has its own limitations. It is often questioned if such techniques are only some kind of brute-force statistical approaches and if they can only work in the context of High Performance Computing with tons of data. Another important question is whether they are really biologically inspired, as claimed in certain cases, and if they can scale well in terms of "intelligence". The dissertation is focused on trying to answer these key questions in the context of Computer Vision and, in particular, Object Recognition, a task that has been heavily revolutionized by recent advances in the field. Practically speaking, these answers are based on an exhaustive comparison between two, very different, deep learning techniques on the aforementioned task: Convolutional Neural Network (CNN) and Hierarchical Temporal memory (HTM). They stand for two different approaches and points of view within the big hat of deep learning and are the best choices to understand and point out strengths and weaknesses of each of them. CNN is considered one of the most classic and powerful supervised methods used today in machine learning and pattern recognition, especially in object recognition. CNNs are well received and accepted by the scientific community and are already deployed in large corporation like Google and Facebook for solving face recognition and image auto-tagging problems. HTM, on the other hand, is known as a new emerging paradigm and a new meanly-unsupervised method, that is more biologically inspired. It tries to gain more insights from the computational neuroscience community in order to incorporate concepts like time, context and attention during the learning process which are typical of the human brain. In the end, the thesis is supposed to prove that in certain cases, with a lower quantity of data, HTM can outperform CNN.

Toll-like receptors in domestic animals

Toll-like receptors are pattern recognition receptors with which hosts recognize pathogen-associated molecular patterns (PAMP). This recognition process is translated rapidly into a meaningful defense reaction. This form of innate host defense is preserved in the animal kingdom: invertebrates heavily depend on it; higher vertebrates also have an adaptive immune system. Both adaptive and innate immune systems are intertwined in that the former also depends on an intact innate recognition and response system. Members of the TLR system cover recognition of parasitic, bacterial or viral germs. Due to the constraints imposed by the necessity to recognize PAMP and to interact with downstream signaling molecules, the TLR system is relatively conserved in evolution. Nevertheless, subtle species differences have been reported for several mammalian TLR members. Examples of this will be given. In all mammalian species investigated, part of the coding sequence is available for the most important TLR members, thus allowing study of expression of these TLR members in various tissues by reverse-transcription polymerase chain reaction in its classical (RT-PCR) and quantitative real time RT-PCR (qRT-PCR) form. In some species, the whole coding sequences of the most important or even all TLR members are known. This allows construction of cDNA and transfection of common host cells, thus permitting functional studies. Extensive investigations were devoted to the study of non-synonymous single nucleotide polymorphisms. In a few cases, expression of a given amino acid in the extracellular (ligand-binding) portion of TLR members could be associated with infectious diseases. This will be discussed below.