Subject independent human action recognition using spatio-depth information and meta-cognitive RBF network

In this paper, we present a machine learning approach for subject independent human action recognition using depth camera, emphasizing the importance of depth in recognition of actions. The proposed approach uses the flow information of all 3 dimensions to classify an action. In our approach, we have obtained the 2-D optical flow and used it along with the depth image to obtain the depth flow (Z motion vectors). The obtained flow captures the dynamics of the actions in space time. Feature vectors are obtained by averaging the 3-D motion over a grid laid over the silhouette in a hierarchical fashion. These hierarchical fine to coarse windows capture the motion dynamics of the object at various scales. The extracted features are used to train a Meta-cognitive Radial Basis Function Network (McRBFN) that uses a Projection Based Learning (PBL) algorithm, referred to as PBL-McRBFN, henceforth. PBL-McRBFN begins with zero hidden neurons and builds the network based on the best human learning strategy, namely, self-regulated learning in a meta-cognitive environment. When a sample is used for learning, PBLMcRBFN uses the sample overlapping conditions, and a projection based learning algorithm to estimate the parameters of the network. The performance of PBL-McRBFN is compared to that of a Support Vector Machine (SVM) and Extreme Learning Machine (ELM) classifiers with representation of every person and action in the training and testing datasets. Performance study shows that PBL-McRBFN outperforms these classifiers in recognizing actions in 3-D. Further, a subject-independent study is conducted by leave-one-subject-out strategy and its generalization performance is tested. It is observed from the subject-independent study that McRBFN is capable of generalizing actions accurately. The performance of the proposed approach is benchmarked with Video Analytics Lab (VAL) dataset and Berkeley Multimodal Human Action Database (MHAD). (C) 2013 Elsevier Ltd. All rights reserved.

Improved subject-independent acoustic-to-articulatory inversion

In subject-independent acoustic-to-articulatory inversion, the articulatory kinematics of a test subject are estimated assuming that the training corpus does not include data from the test subject. The training corpus in subject-independent inversion (SII) is formed with acoustic and articulatory kinematics data and the acoustic mismatch between training and test subjects is then estimated by an acoustic normalization using acoustic data drawn from a large pool of speakers called generic acoustic space (GAS). In this work, we focus on improving the SII performance through better acoustic normalization and adaptation. We propose unsupervised and several supervised ways of clustering GAS for acoustic normalization. We perform an adaptation of acoustic models of GAS using the acoustic data of the training and test subjects in SII. It is found that SII performance significantly improves (similar to 25% relative on average) over the subject-dependent inversion when the acoustic clusters in GAS correspond to phonetic units (or states of 3-state phonetic HMMs) and when the acoustic model built on GAS is adapted to training and test subjects while optimizing the inversion criterion. (C) 2014 Elsevier B.V. All rights reserved.

One Subject, Two Lands: My Journey in Condensed Matter Physics

This is an account of a professional life in the field that was generally known as solid-state physics when I started working in it; India and the United States of America are the countries in which this life was largely played out. My attempts to understand various things in condensed matter physics, and efforts to put together people and activities in India in this field, are mainly the story.

Statistical analysis of synthetic earthquake catalogs generated by models with various levels of fault zone disorder

The stress release model, a stochastic version of the elastic rebound theory, is applied to the large events from four synthetic earthquake catalogs generated by models with various levels of disorder in distribution of fault zone strength (Ben-Zion, 1996) They include models with uniform properties (U), a Parkfield-type asperity (A), fractal brittle properties (F), and multi-size-scale heterogeneities (M). The results show that the degree of regularity or predictability in the assumed fault properties, based on both the Akaike information criterion and simulations, follows the order U, F, A, and M, which is in good agreement with that obtained by pattern recognition techniques applied to the full set of synthetic data. Data simulated from the best fitting stress release models reproduce, both visually and in distributional terms, the main features of the original catalogs. The differences in character and the quality of prediction between the four cases are shown to be dependent on two main aspects: the parameter controlling the sensitivity to departures from the mean stress level and the frequency-magnitude distribution, which differs substantially between the four cases. In particular, it is shown that the predictability of the data is strongly affected by the form of frequency-magnitude distribution, being greatly reduced if a pure Gutenburg-Richter form is assumed to hold out to high magnitudes.

Structures on sliding base subject to horizontal and vertical motions

In this paper particular investigation is directed towards the combined effects of horizontal and vertical motions of real earthquakes to structures resting on sliding base. A simplified method is presented to treat the nonlinear effects of time dependent frictional force of the sliding base as a function of the vertical reaction produced by the foundation. As an example, the El Centro 1940 earthquake record is used on a structural model to show the structural responses due to a sliding base with different frictional and stiffness characteristics. The study shows that vertical ground motion does affect both the superstructure response and the base sliding displacement. Nevertheless, the sliding base isolator is shown to be effective for the reduction of seismic response of a superstructure.

Bibliografía argentina de peces de agua dulce.

Espanol: En la presente lista bibliográfica fueron recopiladas las referencias sobre los peces continentales de la Argentina, del período comprendido entre mediados del siglo XVIII y fines del año 2005. Incluye las listas bibliográficas publicadas durante los años 1981 a 2004, y las citas no mencionadas en ellas. Se incluyeron el ISSN o ISBN según correspondiera, la abreviatura oficial y el lugar de origen de las publicaciones. En algunos casos, los ISSN, las abreviaturas de los nombres de las publicaciones o su procedencia, mencionados en los catálogos, no coinciden con los de la home page de la publicación. Una bibliografía puede ser muy rica y aún estar incompleta. Requiere de sus lectores algún interés histórico, y aún un interés profundo por su tema. Ante una bibliografía, muchos investigadores preferirían no encontrar algunas referencias, y de hecho, muchas son oportunamente olvidadas. Por no saber como hacerlo, o por menosprecio, estas listas raramente son citadas en los trabajos, aunque sobre algunos temas en particular, sería realmente difícil formarse una idea si las bibliografías no existieran. Aún desde el comienzo es complicado precisar un criterio de inclusiones. Por ejemplo, gran parte de la ictiofauna Argentina se encuentra también en Brasil. ¿Justifica esto incluir informes perdidos sobre artes de pesca en una cuenca distante? ¿Deben los clásicos, que todo el mundo conoce y el que se inicia encontrará sin dificultad, ser incluidos? Aún a un grupo que se dedicara full time a este trabajo, le sería difícil verificar la precisión de las citas antiguas, en las que fechas y autoría cambian según la investigación histórica. En una bibliografía más o menos general, la perfección atenta contra la publicación. Sin embargo, pensamos que es conveniente hacerlas. Una mirada a este volumen, muestra la enorme cantidad de desarrollos en muchos temas, y la regla que uno de nosotros ha mencionado desde hace tiempo: siempre hay más publicado sobre un tema de lo que uno cree. La sospecha de que con sólo mirar lo que está hecho, muchos subsidios podrían utilizarse para algo más útil que algunas evaluaciones repetidas de recursos o biodiversidad, es un poco pesimista y no haremos perder trabajos insistiendo en eso. Cada generación elige sus metas, su propia base epistemológica, sus trabajos preferidos y los que desecha. Aún en trabajos perdidos o de mala calidad, es posible encontrar datos valiosos. Ningún proyecto, por mejor diseñado que esté, podrá mostrar en el presente los organismos que vivían en el pasado en un lugar en el que las condiciones han cambiado, o lo hará en términos de otra disciplina. En los temas aplicados la información del pasado puede ser importantísima. Aún en una disciplina tan conservadora como la nomenclatura, los cambios pueden ser exasperantes; no pueden serlo menos en las que intrínsecamente, como la ecología, es lo que estudian. Para dar una idea más precisa del desarrollo de la ictiología en la Argentina, esta lista podría ir acompañada de una apreciación crítica. Entendemos que una tarea así exige un trabajo diferente, de cierta magnitud y con no pocos elementos históricos. Aunque tiene deficiencias, la ictiología argentina constituye una acumulación de conocimientos de considerable calidad y pertinencia para la historia natural de América del Sur. Dejamos a los lectores que cada uno haga la suya. English: For the present list, references on freshwater fishes of Argentina were compiled from the period between middle XVIII Century and the end of the year 2005. It includes previous lists published during 1981 to 2004, and references not mentioned therein. The ISSN o ISBN numbers were included, as well as the official abbreviations and the place of origin of the periodicals. In some cases, these data as quoted in catalogs, do not agree with those in the home page of the publication. A bibliography may be very rich and anyway never complete. It requires from its readers some historic interest and indeed a deep interest on his (her) subject. Browsing a bibliography, many researchers would prefer not to find some references, and in fact, sometimes they forget some of them. Not imagining how to do it, or because people do not concede importance to them, bibliographic lists are rarely quoted in papers, though some subjects would be rather difficult to understand if list of publications would not exist. Even from the beginning, it is difficult to precise a criterion of inclusions. For example, many Argentine fishes occur also in Brazil. Does this justify the inclusion of grey reports on a distant basin? Should classic works, that everybody knows and are easily found, be included? Is near impossible, even for a group dedicated full time to this work, to verify the precision of old citations, whose dates and authorship change according to authorities and historical research. In a more or less general bibliography, completeness is against publication. Nevertheless, we think that is convenient to prepare these lists. A look at this volume shows the enormous developments in many subjects, and the rule that one of us mentioned long ago: there are always more papers on any subject than one suspects. Looking at what has already been done raises the suspicion that many grants could be used for something more useful than repeated evaluations of biodiversity or resources. This is a bit pessimistic, and we do not want to erase working opportunities. Each generation chooses its targets, its own epistemological base, its preferred papers and those that rejects. Even in lost or bad quality papers, the possibility of finding valuable information exists. No project, whatever the appropriateness of its design, could show at present which organisms lived in the past in a place where environmental conditions have changed, or it will do it in terms of another discipline. In applied subjects, information from the past can be very important. Even in a conservative discipline as nomenclature, changes can be exasperating. They are not lesser in those like ecology, where change itself is studied. To provide a more precise idea of the development of ichthyology in Argentina, this list could be accompanied by a critical appreciation. We understood that such an aim requires a different work, with no few historical elements and of certain magnitude. In spite of some deficiencies, Argentine ichthyology, resulting from collaboration of both local and foreign people, constitutes a bulk of knowledge of considerable quality and pertinence for the natural history of South America. We leave each reader to make his (or her) own evaluation. (Texto en Espanol. PDF tiene cien setenta paginas.)

Asymptotic Hyperstability of a Class of Linear Systems under Impulsive Controls Subject to an Integral Popovian Constraint

This paper is focused on the study of the important property of the asymptotic hyperstability of a class of continuous-time dynamic systems. The presence of a parallel connection of a strictly stable subsystem to an asymptotically hyperstable one in the feed-forward loop is allowed while it has also admitted the generation of a finite or infinite number of impulsive control actions which can be combined with a general form of nonimpulsive controls. The asymptotic hyperstability property is guaranteed under a set of sufficiency-type conditions for the impulsive controls.