Robust Multimodal Person Identification With Limited Training Data

This paper presents a novel method of audio-visual feature-level fusion for person identification where both the speech and facial modalities may be corrupted, and there is a lack of prior knowledge about the corruption. Furthermore, we assume there are limited amount of training data for each modality (e.g., a short training speech segment and a single training facial image for each person). A new multimodal feature representation and a modified cosine similarity are introduced to combine and compare bimodal features with limited training data, as well as vastly differing data rates and feature sizes. Optimal feature selection and multicondition training are used to reduce the mismatch between training and testing, thereby making the system robust to unknown bimodal corruption. Experiments have been carried out on a bimodal dataset created from the SPIDRE speaker recognition database and AR face recognition database with variable noise corruption of speech and occlusion in the face images. The system's speaker identification performance on the SPIDRE database, and facial identification performance on the AR database, is comparable with the literature. Combining both modalities using the new method of multimodal fusion leads to significantly improved accuracy over the unimodal systems, even when both modalities have been corrupted. The new method also shows improved identification accuracy compared with the bimodal systems based on multicondition model training or missing-feature decoding alone.

Attention-based modulation of tactile stimuli: A comparison between prefrontal lesion patients and healthy age-matched controls

Objectives: To investigate the role of the prefrontal cortex in attention-based modulation of cortical somatosensory processing.

Methods: Six prefrontal stroke patients were compared with eleven neurologically intact older adults during a vibrotactile discrimination task. All subjects attended to stimuli on one digit while ignoring distracter stimuli on a separate digit of the same hand. Subjects were required to report infrequent targets on the attended digit only. Throughout testing electroencephalography was used to measure event-related potentials for both task-relevant and irrelevant stimuli.

Results: Prefrontal patients demonstrated significant changes in cortical somatosensory processing based on attention compared to age-matched controls. This was evident both in early unimodal somatosensory processing (i.e. P100) and in later cortical processing stages (i.e. long-latency positivity). Moreover, there was a tendency towards a tonic loss of inhibition over early somatosensory cortical processing (i.e. P50).

Conclusions: The attention-based modulation noted for neurologically intact older adults was absent in prefrontal lesion patients.

Significance: The present study highlights the important role of prefrontal regions in sustaining inhibition over early sensory cortical processing stages and in modifying somatosensory transmission based on task-relevance. Notably these deficits extend beyond those previously shown to occur as a function of age.

Transform both sides model: A parametric approach

A parametric regression model for right-censored data with a log-linear median regression function and a transformation in both response and regression parts, named parametric Transform-Both-Sides (TBS) model, is presented. The TBS model has a parameter that handles data asymmetry while allowing various different distributions for the error, as long as they are unimodal symmetric distributions centered at zero. The discussion is focused on the estimation procedure with five important error distributions (normal, double-exponential, Student's t, Cauchy and logistic) and presents properties, associated functions (that is, survival and hazard functions) and estimation methods based on maximum likelihood and on the Bayesian paradigm. These procedures are implemented in TBSSurvival, an open-source fully documented R package. The use of the package is illustrated and the performance of the model is analyzed using both simulated and real data sets.

A new statistical framework for the quantification of covariate associations with species distributions

Identifying processes that shape species geographical ranges is a prerequisite for understanding environmental change. Currently, species distribution modelling methods do not offer credible statistical tests of the relative influence of climate factors and typically ignore other processes (e.g. biotic interactions and dispersal limitation). We use a hierarchical model fitted with Markov Chain Monte Carlo to combine ecologically plausible niche structures using regression splines to describe unimodal but potentially skewed response terms. We apply spatially explicit error terms that account for (and may help identify) missing variables. Using three example distributions of European bird species, we map model results to show sensitivity to change in each covariate. We show that the overall strength of climatic association differs between species and that each species has considerable spatial variation in both the strength of the climatic association and the sensitivity to climate change. Our methods are widely applicable to many species distribution modelling problems and enable accurate assessment of the statistical importance of biotic and abiotic influences on distributions.

Impact of biodiversity loss on production in complex marine food webs mitigated by prey-release

Public concern over biodiversity loss is often rationalized as a threat to ecosystem functioning, but biodiversity-ecosystem functioning (BEF) relations are hard to empirically quantify at large scales. We use a realistic marine food-web model, resolving species over five trophic levels, to study how total fish production changes with species richness. This complex model predicts that BEF relations, on average, follow simple Michaelis-Menten curves when species are randomly deleted. These are shaped mainly by release of fish from predation, rather than the release from competition expected from simpler communities. Ordering species deletions by decreasing body mass or trophic level, representing 'fishing down the food web', accentuates prey-release effects and results in unimodal relationships. In contrast, simultaneous unselective harvesting diminishes these effects and produces an almost linear BEF relation, with maximum multispecies fisheries yield at approximate to 40% of initial species richness. These findings have important implications for the valuation of marine biodiversity.

Evolutionary branching in deme-structured populations.

Adaptive dynamics shows that a continuous trait under frequency dependent selection may first converge to a singular point followed by spontaneous transition from a unimodal trait distribution into a bimodal one, which is called "evolutionary branching". Here, we study evolutionary branching in a deme-structured population by constructing a quantitative genetic model for the trait variance dynamics, which allows us to obtain an analytic condition for evolutionary branching. This is first shown to agree with previous conditions for branching expressed in terms of relatedness between interacting individuals within demes and obtained from mutant-resident systems. We then show this branching condition can be markedly simplified when the evolving trait affect fecundity and/or survival, as opposed to affecting population structure, which would occur in the case of the evolution of dispersal. As an application of our model, we evaluate the threshold migration rate below which evolutionary branching cannot occur in a pairwise interaction game. This agrees very well with the individual-based simulation results.

Multimodal carrier liability in the U.S. and Canada : towards uniformity of applicable rules?

"Thèse présentée à la Faculté des études supérieures en vue de l'obtention du grade de Docteur en Droit (LL.D.) et à la Faculté de Droit et de Sciences Politiques de l'Université de Nantes en vue de l'obtention du grade de Docteur"

Polyamides and polyesters made of bile acids in the main chain

La préparation de polymères à base d’acides biliaires, molécules biologiques, a attiré l'attention des chercheurs en raison des applications potentielles dans les domaines biomédicaux et pharmaceutiques. L’objectif de ce travail est de synthétiser de nouveaux biopolymères dont la chaîne principale est constituée d’unités d’acides biliaires. La polymérisation par étapes a été adoptée dans ce projet afin de préparer les deux principales classes de polymères utilisés en fibres textiles: les polyamides et les polyesters. Des monomères hétéro-fonctionnels à base d’acides biliaires ont été synthétisés et utilisés afin de surmonter le déséquilibre stoechiométrique lors de la polymérisation par étapes. Le dérivé de l’acide lithocholique modifié par une fonction amine et un groupement carboxylique protégé a été polymérisé en masse à températures élevées. Les polyamides obtenus sont très peu solubles dans les solvants organiques. Des polyamides et des polyesters solubles en milieu organique ont pu être obtenus dans des conditions modérées en utilisant l’acide cholique modifié par des groupements azide et alcyne. La polymérisation a été réalisée par cycloaddition azoture-alcyne catalysée par l'intermédiaire du cuivre(Ι) avec deux systèmes catalytiques différents, le bromure de cuivre(I) et le sulfate de cuivre(II). Seul le bromure de cuivre(Ι) s’est avéré être un catalyseur efficace pour le système, permettant la préparation des polymères avec un degré de polymérisation égale à 50 et une distribution monomodale de masse moléculaire (PDI ˂ 1.7). Les polymères synthétisés à base d'acide cholique sont thermiquement stables (307 °C ≤ Td ≤ 372 °C) avec des températures de transition vitreuse élevées (137 °C ≤ Tg ≤ 167 °C) et modules de Young au-dessus de 280 MPa, dépendamment de la nature chimique du lien.

Impacts of Climate variability on Agriculture in Kerala

The study revealed that southwest monsoon rainfall in Kerala has been declining while increasing in post monsoon season. The annual rainfall exhibits a cyclic trend of 40-60 years, with a significant decline in recent decades. The intensity of climatological droughts was increasing across the State of Kerala through it falls under heavy rainfall zone due to unimodal rainfall pattern. The moisture index across the State of Kerala was moving from B4 to B3 humid, indicating that the State was moving from wetness to dryness within the humid climate.The study confirms that a warming Kerala is real as maximum, minimum and mean temperatures and temperature ranges are increasing. The rate of increase in maximum temperature was high (1.46°C) across the high ranges, followed by the coastal belt (1.09°C) of Kerala while the rate of increase was relatively marginal (0.25°C) across the midlands. The rate of increase in temperature across the high ranges is probably high because of deforestation. It indicates that the highranges and coastal belts in Kerala are vulnerable to global warming and climate change when compared to midlands.Interestingly, the trend in annual rainfall is increasing at Pampadumpara (Idukki), while declining at Ambalavayal across the highranges. In the case of maximum temperature, it was showing increasing trend at Pampadumpara while declining trend at Ambalavayal. In the case of minimum temperature it is declining at Pampadumpara while increasing in Ambalavalal.The paddy productivity in Kerala during kharif / virippu is unlikely to decline due to increasing temperature on the basis of long term climate change, but likely to decline to a considerable extent due to prolonged monsoon season, followed by unusual summer rains as noticed in 2007-08 and 2010-11.All the plantation crops under study are vulnerable to climate variability such as floods and droughts rather than long term changes in temperature and rainfall.

Spatial and temporal characteristics of rain intensity in the peninsular Malaysia using TRMM rain rate

The present study is focused on the intensity distribution of rainfall in different classes and their contribution to the total seasonal rainfall. In addition, we studied the spatial and diurnal variation of the rainfall in the study areas. For the present study, we retrieved data from TRMM (Tropical Rain Measuring Mission) rain rate available in every 3 h temporal and 25 km spatial resolutions. Moreover, station rainfall data is used to validate the TRMM rain rate and found significant correlation between them (linear correlation coefficients are 0.96, 0.85, 0.75 and 0.63 for the stations Kota Bharu, Senai, Cameron highlands and KLIA, respectively). We selected four areas in the Peninsular Malaysia and they are south coastal, east coastal, west coastal and highland regions. Diurnal variation of frequency of rain occurrence is different for different locations. We noticed bimodal variation in the coastal areas in most of the seasons and unimodal variation in the highland/inland area. During the southwest monsoon period in the west coastal stations, there is no distinct diurnal variation. The distribution of different intensity classes during different seasons are explained in detail in the results

A Prototype for a Multimodal Biometric Security system based on Face and Audio Signatures

Any automatically measurable, robust and distinctive physical characteristic or personal trait that can be used to identify an individual or verify the claimed identity of an individual, referred to as biometrics, has gained significant interest in the wake of heightened concerns about security and rapid advancements in networking, communication and mobility. Multimodal biometrics is expected to be ultra-secure and reliable, due to the presence of multiple and independent—verification clues. In this study, a multimodal biometric system utilising audio and facial signatures has been implemented and error analysis has been carried out. A total of one thousand face images and 250 sound tracks of 50 users are used for training the proposed system. To account for the attempts of the unregistered signatures data of 25 new users are tested. The short term spectral features were extracted from the sound data and Vector Quantization was done using K-means algorithm. Face images are identified based on Eigen face approach using Principal Component Analysis. The success rate of multimodal system using speech and face is higher when compared to individual unimodal recognition systems

Fronts from complex two-dimensional dispersal kernels: theory and application to Reid's paradox

Bimodal dispersal probability distributions with characteristic distances differing by several orders of magnitude have been derived and favorably compared to observations by Nathan [Nature (London) 418, 409 (2002)]. For such bimodal kernels, we show that two-dimensional molecular dynamics computer simulations are unable to yield accurate front speeds. Analytically, the usual continuous-space random walks (CSRWs) are applied to two dimensions. We also introduce discrete-space random walks and use them to check the CSRW results (because of the inefficiency of the numerical simulations). The physical results reported are shown to predict front speeds high enough to possibly explain Reid's paradox of rapid tree migration. We also show that, for a time-ordered evolution equation, fronts are always slower in two dimensions than in one dimension and that this difference is important both for unimodal and for bimodal kernels

Factors influencing zooplankton size structure at contrasting temperatures in coastal shallow lakes: Implications for effects of climate change

We assessed the importance of temperature, salinity, and predation for the size structure of zooplankton and provided insight into the future ecological structure and function of shallow lakes in a warmer climate. Artificial plants were introduced in eight comparable coastal shallow brackish lakes located at two contrasting temperatures: cold-temperate and Mediterranean climate region. Zooplankton, fish, and macroinvertebrates were sampled within the plants and at open-water habitats. The fish communities of these brackish lakes were characterized by small-sized individuals, highly associated with submerged plants. Overall, higher densities of small planktivorous fish were recorded in the Mediterranean compared to the cold-temperate region, likely reflecting temperature-related differences as have been observed in freshwater lakes. Our results suggest that fish predation is the major control of zooplankton size structure in brackish lakes, since fish density was related to a decrease in mean body size and density of zooplankton and this was reflected in a unimodal shaped biomass-size spectrum with dominance of small sizes and low size diversity. Salinity might play a more indirect role by shaping zooplankton communities toward more salt-tolerant species. In a global-warming perspective, these results suggest that changes in the trophic structure of shallow lakes in temperate regions might be expected as a result of the warmer temperatures and the potentially associated increases in salinity. The decrease in the density of largebodied zooplankton might reduce the grazing on phytoplankton and thus the chances of maintaining the clear water state in these ecosystems

State estimation using the particle filter with mode tracking

A particle filter is a data assimilation scheme that employs a fully nonlinear, non-Gaussian analysis step. Unfortunately as the size of the state grows the number of ensemble members required for the particle filter to converge to the true solution increases exponentially. To overcome this Vaswani [Vaswani N. 2008. IEEE Trans Signal Process 56:4583–97] proposed a new method known as mode tracking to improve the efficiency of the particle filter. When mode tracking, the state is split into two subspaces. One subspace is forecast using the particle filter, the other is treated so that its values are set equal to the mode of the marginal pdf. There are many ways to split the state. One hypothesis is that the best results should be obtained from the particle filter with mode tracking when we mode track the maximum number of unimodal dimensions. The aim of this paper is to test this hypothesis using the three dimensional stochastic Lorenz equations with direct observations. It is found that mode tracking the maximum number of unimodal dimensions does not always provide the best result. The best choice of states to mode track depends on the number of particles used and the accuracy and frequency of the observations.

Axonal velocity distributions in neural feld equations

By modelling the average activity of large neuronal populations, continuum mean field models (MFMs) have become an increasingly important theoretical tool for understanding the emergent activity of cortical tissue. In order to be computationally tractable, long-range propagation of activity in MFMs is often approximated with partial differential equations (PDEs). However, PDE approximations in current use correspond to underlying axonal velocity distributions incompatible with experimental measurements. In order to rectify this deficiency, we here introduce novel propagation PDEs that give rise to smooth unimodal distributions of axonal conduction velocities. We also argue that velocities estimated from fibre diameters in slice and from latency measurements, respectively, relate quite differently to such distributions, a significant point for any phenomenological description. Our PDEs are then successfully fit to fibre diameter data from human corpus callosum and rat subcortical white matter. This allows for the first time to simulate long-range conduction in the mammalian brain with realistic, convenient PDEs. Furthermore, the obtained results suggest that the propagation of activity in rat and human differs significantly beyond mere scaling. The dynamical consequences of our new formulation are investigated in the context of a well known neural field model. On the basis of Turing instability analyses, we conclude that pattern formation is more easily initiated using our more realistic propagator. By increasing characteristic conduction velocities, a smooth transition can occur from self-sustaining bulk oscillations to travelling waves of various wavelengths, which may influence axonal growth during development. Our analytic results are also corroborated numerically using simulations on a large spatial grid. Thus we provide here a comprehensive analysis of empirically constrained activity propagation in the context of MFMs, which will allow more realistic studies of mammalian brain activity in the future.