Identity and gender recognition using the ENCARA real-time face detector

[EN]This paper presents recognition results based on a PCA representation and classification with SVMs and temporal coherence.

ENCARA: real-time detection of frontal faces

[EN]This paper describes a real-time approach for face detection and selection of frontal views, for further processing. Typically, face detection papers provide results for a set of single images but the problem of face detection in video streams rarely is tackled. Instead of performing an exhaustive search for every video stream frame a set of opportunistic ideas applied in a cascade fashion and based on temporal and spatial coherence provide promising results in real-time.

A C80 DSP-based active vision system for real-time tracking

[EN]Active Vision Systems can be considered as dynamical systems which close the loop around artificial visual perception, controlling camera parameters, motion and also controlling processing to simplify, accelerate and do more robust visual perception. Research and Development in Active Vision Systems [Aloi87], [Bajc88] is a main area of interest in Computer Vision, mainly by its potential application in different scenarios where real-time performance is needed such as robot navigation, surveillance, visual inspection, among many others. Several systems have been developed during last years using robotic-heads for this purpose...

Using the interaural time difference and cross-correlation to localise short-term complex noises

The mammalian binaural cue of interaural time difference (ITD) and cross-correlation have long been used to determine the point of origin of a sound source. The ITD can be defined as the different points in time at which a sound from a single location arrives at each individual ear [1]. From this time difference, the brain can calculate the angle of the sound source in relation to the head [2]. Cross-correlation compares the similarity of each channel of a binaural waveform producing the time lag or offset required for both channels to be in phase with one another. This offset corresponds to the maximum value produced by the cross-correlation function and can be used to determine the ITD and thus the azimuthal angle θ of the original sound source. However, in indoor environments, cross-correlation has been known to have problems with both sound reflections and reverberations. Additionally, cross-correlation has difficulties with localising short-term complex noises when they occur during a longer duration waveform, i.e. in the presence of background noise. The crosscorrelation algorithm processes the entire waveform and the short-term complex noise can be ignored. This paper presents a technique using thresholding which enables higher-localisation abilities for short-term complex sounds in the midst of background noise. To determine the success of this thresholding technique, twenty-five sounds were recorded in a dynamic and echoic environment. The twenty-five sounds consist of hand-claps, finger-clicks and speech. The proposed technique was compared to the regular cross-correlation function for the same waveforms, and an average of the azimuthal angles determined for each individual sample. The sound localisation ability for all twenty-five sound samples is as follows: average of the sampled angles using cross-correlation: 44%; cross-correlation technique with thresholding: 84%. From these results, it is clear that this proposed technique is very successful for the localisation of short-term complex sounds in the midst of background noise and in a dynamic and echoic indoor environment.

Survivorship and Clinical Outcome of the Minimally Invasive Uniglide Medial Fixed Bearing, All-polyethylene Tibia, Unicompartmental Knee Arthroplasty at a Mean Follow-up of 7.3 Years

Background: Medial UKA performed in England and Wales represents 7 to 11% of all knee arthroplasty procedures, and is most commonly performed using mobile-bearing designs. Fixed bearing eliminates the risk of bearing dislocation, however some studies have shown higher revision rates for all-polyethylene tibial components compared to those that utilize metal-backed implants. The aim of the study is to analyse survivorship and maximum 8-year clinical outcome of medial fixed bearing, Uniglide unicompartmental knee arthroplasty performed using an all-polyethylene tibial component with a minimal invasive approach. Methods: Between 2002 and 2009, 270 medial fixed UKAs were performed in our unit. Patients were reviewed pre-operatively, 5 and 8 years post-operatively. Clinical and radiographic reviews were carried out. Patients’ outcome scores (Oxford, WOMAC and American Knee Score) were documented in our database and analysed. Results: Survival and clinical outcome data of 236 knees with a mean 7.3 years follow-up are reported. Every patient with less than 4.93 years follow-up underwent a revision. The patients’ average age at the time of surgery was 69.5 years. The American Knee Society Pain and Function scores, the Oxford Knee Score and the WOMAC score all improved significantly. The 5 years survival rate was 94.1% with implant revision surgery as an end point. The estimated 10 years survival rate is 91.3%. 14 patients were revised before the 5 year follow-up. Conclusion: Fixed bearing Uniglide UKA with an all-polyethylene tibial component is a valuable tool in the management of a medial compartment osteoarthritis, affording good short term survivorship.

Asymptotic Change-Point Analysis of the Dependencies in Time Series

In recent papers, Wied and his coauthors have introduced change-point procedures to detect and estimate structural breaks in the correlation between time series. To prove the asymptotic distribution of the test statistic and stopping time as well as the change-point estimation rate, they use an extended functional Delta method and assume nearly constant expectations and variances of the time series. In this thesis, we allow asymptotically infinitely many structural breaks in the means and variances of the time series. For this setting, we present test statistics and stopping times which are used to determine whether or not the correlation between two time series is and stays constant, respectively. Additionally, we consider estimates for change-points in the correlations. The employed nonparametric statistics depend on the means and variances. These (nuisance) parameters are replaced by estimates in the course of this thesis. We avoid assuming a fixed form of these estimates but rather we use "blackbox" estimates, i.e. we derive results under assumptions that these estimates fulfill. These results are supplement with examples. This thesis is organized in seven sections. In Section 1, we motivate the issue and present the mathematical model. In Section 2, we consider a posteriori and sequential testing procedures, and investigate convergence rates for change-point estimation, always assuming that the means and the variances of the time series are known. In the following sections, the assumptions of known means and variances are relaxed. In Section 3, we present the assumptions for the mean and variance estimates that we will use for the mean in Section 4, for the variance in Section 5, and for both parameters in Section 6. Finally, in Section 7, a simulation study illustrates the finite sample behaviors of some testing procedures and estimates.

A.G.B. ter Meulen, Representing Time in Natural Language: The Dynamic Interpretation of Tense and Aspect, The MIT Press, 1997, ISBN: 978-0262200998 [Book review]

Abstract not available

Artificial Skin - Culturing of Different Skin Cell Lines for Generating an Artificial Skin Substitute on Cross-Weaved Spider Silk Fibres

Background: In the field of Plastic Reconstructive Surgery the development of new innovative matrices for skin repair is in urgent need. The ideal biomaterial should promote attachment, proliferation and growth of cells. Additionally, it should degrade in an appropriate time period without releasing harmful substances, but not exert a pathological immune response. Spider dragline silk from Nephila spp meets these demands to a large extent. Methodology/Principal Findings: Native spider dragline silk, harvested directly out of Nephila spp spiders, was woven on steel frames. Constructs were sterilized and seeded with fibroblasts. After two weeks of cultivating single fibroblasts, keratinocytes were added to generate a bilayered skin model, consisting of dermis and epidermis equivalents. For the next three weeks, constructs in co-culture were lifted on an originally designed setup for air/liquid interface cultivation. After the culturing period, constructs were embedded in paraffin with an especially developed program for spidersilk to avoid supercontraction. Paraffin cross-sections were stained in Haematoxylin & Eosin (H&E) for microscopic analyses. Conclusion/Significance: Native spider dragline silk woven on steel frames provides a suitable matrix for 3 dimensional skin cell culturing. Both fibroblasts and keratinocytes cell lines adhere to the spider silk fibres and proliferate. Guided by the spider silk fibres, they sprout into the meshes and reach confluence in at most one week. A well-balanced, bilayered cocultivation in two continuously separated strata can be achieved by serum reduction, changing the medium conditions and the cultivation period at the air/liquid interphase. Therefore spider silk appears to be a promising biomaterial for the enhancement of skin regeneration.

'Investigating Artificial Immune Systems For Job Shop Rescheduling In Changing Environments'

ABSTRACT Artificial immune system can be used to generate schedules in changing environments and it has been proven to be more robust than schedules developed using a genetic algorithm. Good schedules can be produced especially when the number of the antigens is increased. However, an increase in the range of the antigens had somehow affected the fitness of the immune system. In this research, we are trying to improve the result of the system by rescheduling the same problem using the same method while at the same time maintaining the robustness of the schedules.

'Artificial Immune Systems: A new Decision Making Paradigm for Operational Research'

Over the last few years, more and more heuristic decision making techniques have been inspired by nature, e.g. evolutionary algorithms, ant colony optimisation and simulated annealing. More recently, a novel computational intelligence technique inspired by immunology has emerged, called Artificial Immune Systems (AIS). This immune system inspired technique has already been useful in solving some computational problems. In this keynote, we will very briefly describe the immune system metaphors that are relevant to AIS. We will then give some illustrative real-world problems suitable for AIS use and show a step-by-step algorithm walkthrough. A comparison of AIS to other well-known algorithms and areas for future work will round this keynote off. It should be noted that as AIS is still a young and evolving field, there is not yet a fixed algorithm template and hence actual implementations might differ somewhat from the examples given here

Efeitos da tilápia do Nilo (Oreochromis niloticus) e do enriquecimento por nutrientes sobre a comunidade planctônica em um lago artificial no semi-árido brasileiro

The major aim of this study was to test the hypothesis that the introduction of the Nile tilapia (Oreochromis niloticus) and the enrichment with nutrients (N and P) interact synergistically to change the structure of plankton communities, increase phytoplankton biomass and decrease water transparency of a semi-arid tropical reservoir. One field experiment was performed during five weeks in twenty enclosures (8m3) to where four treatments were randomly allocated: with tilapia addition (T), with nutrients addition (NP), with tilapia and nutrients addition (T+NP) and a control treatment with no tilapia or nutrients addition (C). A two-way repeated measures ANOVA was done to test for time (t), tilapia (T) and nutrient (NP) effects and their interaction on water transparency, total phosphorus, total nitrogen, phytoplankton and zooplankton. The results show that there was no effect of nutrient addition on these variables but significant fish effects on the biomass of total zooplankton, nauplii, rotifers, cladocerans and calanoid copepods, on the biovolume of Bacillariophyta, Zygnemaphyceae and large algae (GALD ≥ 50 μm) and on Secchi depth. In addition, we found significant interaction effects between tilapia and nutrients on Secchi depth and rotifers. Overall, tilapia decreased the biomass of most zooplankton taxa and large algae (diatoms) and decreased the water transparency while nutrient enrichment increased the biomass of zooplankton (rotifers) but only in the absence of tilapia. In conclusion, the influence of fish on the reservoir plankton community and water transparency was greater than that of nutrient loading. This finding suggests that biomanipulation should be a greater priority in the restoration of eutrophic reservoirs in tropical semi-arid regions

Comparing fixed and flexible stocking as adaptations to inter-annual rainfall variability in the extensive beef industry of northern Australia

Many beef producers within the extensive cattle industry of northern Australia attempt to maintain a constant herd size from year-to-year (fixed stocking), whereas others adjust stock numbers to varying degrees annually in response to changes in forage supply. The effects of these strategies on pasture condition and cattle productivity cannot easily be assessed by grazing trials. Simulation studies, which include feedbacks of changes to pasture condition on cattle liveweight gain, can extend the results of grazing trials both spatially and temporally. They can compare a large number of strategies, over long periods of time, for a range of climate periods, at locations which differ markedly in climate. This simulation study compared the pasture condition and cattle productivity achieved by fixed stocking at the long-term carrying capacity with that of 55 flexible stocking strategies at 28 locations across Queensland and the Northern Territory. Flexible stocking strategies differed markedly in the degree they increased or decreased cattle stocking rates after good and poor pasture growing seasons, respectively. The 28 locations covered the full range in average annual rainfall and inter-annual rainfall variability experienced across northern Australia. Constrained flexibility, which limited increases in stocking rates after good growing seasons to 10% but decreased them by up to 20% after poor growing seasons, provides sustainable productivity gains for cattle producers in northern Australia. This strategy can improve pasture condition and increase cattle productivity relative to fixed stocking at the long-term carrying capacity, and its capacity to do this was greatest in the semiarid rangeland regions that contain the majority of beef cattle in northern Australia. More flexible stocking strategies, which also increased stocking rates after good growing seasons by only half as much as they decreased them after poor growing seasons, were equally sustainable and more productive than constrained flexibility, but are often impractical at property and industry scales. Strategies with the highest limits (e.g. 70%) for both annual increases and decreases in stocking rates could achieve higher cattle productivity, but this was at the expense of pasture condition and was not sustainable. Constrained flexible stocking, with a 10% limit for increases and a 20% limit for decreases in stocking rates annually, is a risk-averse adaptation to high and unpredictable rainfall variability for the extensive beef industry of northern Australia. © Australian Rangeland Society 2016.

Dendritic Cells for Real-Time Anomaly Detection

Dendritic Cells (DCs) are innate immune system cells which have the power to activate or suppress the immune system. The behaviour of human DCs is abstracted to form an algorithm suitable for anomaly detection. We test this algorithm on the real-time problem of port scan detection. Our results show a significant difference in artificial DC behaviour for an outgoing portscan when compared to behaviour for normal processes.

'Investigating Artificial Immune Systems For Job Shop Rescheduling In Changing Environments'

ABSTRACT Artificial immune system can be used to generate schedules in changing environments and it has been proven to be more robust than schedules developed using a genetic algorithm. Good schedules can be produced especially when the number of the antigens is increased. However, an increase in the range of the antigens had somehow affected the fitness of the immune system. In this research, we are trying to improve the result of the system by rescheduling the same problem using the same method while at the same time maintaining the robustness of the schedules.

SEQUENTIAL DECISIONS AND PREDICTIONS IN NATURAL LANGUAGE PROCESSING

Natural language processing has achieved great success in a wide range of ap- plications, producing both commercial language services and open-source language tools. However, most methods take a static or batch approach, assuming that the model has all information it needs and makes a one-time prediction. In this disser- tation, we study dynamic problems where the input comes in a sequence instead of all at once, and the output must be produced while the input is arriving. In these problems, predictions are often made based only on partial information. We see this dynamic setting in many real-time, interactive applications. These problems usually involve a trade-off between the amount of input received (cost) and the quality of the output prediction (accuracy). Therefore, the evaluation considers both objectives (e.g., plotting a Pareto curve). Our goal is to develop a formal understanding of sequential prediction and decision-making problems in natural language processing and to propose efficient solutions. Toward this end, we present meta-algorithms that take an existent batch model and produce a dynamic model to handle sequential inputs and outputs. Webuild our framework upon theories of Markov Decision Process (MDP), which allows learning to trade off competing objectives in a principled way. The main machine learning techniques we use are from imitation learning and reinforcement learning, and we advance current techniques to tackle problems arising in our settings. We evaluate our algorithm on a variety of applications, including dependency parsing, machine translation, and question answering. We show that our approach achieves a better cost-accuracy trade-off than the batch approach and heuristic-based decision- making approaches. We first propose a general framework for cost-sensitive prediction, where dif- ferent parts of the input come at different costs. We formulate a decision-making process that selects pieces of the input sequentially, and the selection is adaptive to each instance. Our approach is evaluated on both standard classification tasks and a structured prediction task (dependency parsing). We show that it achieves similar prediction quality to methods that use all input, while inducing a much smaller cost. Next, we extend the framework to problems where the input is revealed incremen- tally in a fixed order. We study two applications: simultaneous machine translation and quiz bowl (incremental text classification). We discuss challenges in this set- ting and show that adding domain knowledge eases the decision-making problem. A central theme throughout the chapters is an MDP formulation of a challenging problem with sequential input/output and trade-off decisions, accompanied by a learning algorithm that solves the MDP.