Anchored deformable face ensemble alignment

At present, many approaches have been proposed for deformable face alignment with varying degrees of success. However, the common drawback to nearly all these approaches is the inaccurate landmark registrations. The registration errors which occur are predominantly heterogeneous (i.e. low error for some frames in a sequence and higher error for others). In this paper we propose an approach for simultaneously aligning an ensemble of deformable face images stemming from the same subject given noisy heterogeneous landmark estimates. We propose that these initial noisy landmark estimates can be used as an “anchor” in conjunction with known state-of-the-art objectives for unsupervised image ensemble alignment. Impressive alignment performance is obtained using well known deformable face fitting algorithms as “anchors.

Ensemble-based DDoS detection and mitigation model

This work-in-progress paper presents an ensemble-based model for detecting and mitigating Distributed Denial-of-Service (DDoS) attacks, and its partial implementation. The model utilises network traffic analysis and MIB (Management Information Base) server load analysis features for detecting a wide range of network and application layer DDoS attacks and distinguishing them from Flash Events. The proposed model will be evaluated against realistic synthetic network traffic generated using a software-based traffic generator that we have developed as part of this research. In this paper, we summarise our previous work, highlight the current work being undertaken along with preliminary results obtained and outline the future directions of our work.

"Who Are You" by DeepBlue Orchestra - Production of Recordings

The development and recording of 10 songs for a CD to accompany DeepBlue's new live orchestra production "Who Are You" which began touring Australia and Asia in 2012.

An ensemble method for predicting subnuclear localizations from primary protein structures

Background Predicting protein subnuclear localization is a challenging problem. Some previous works based on non-sequence information including Gene Ontology annotations and kernel fusion have respective limitations. The aim of this work is twofold: one is to propose a novel individual feature extraction method; another is to develop an ensemble method to improve prediction performance using comprehensive information represented in the form of high dimensional feature vector obtained by 11 feature extraction methods. Methodology/Principal Findings A novel two-stage multiclass support vector machine is proposed to predict protein subnuclear localizations. It only considers those feature extraction methods based on amino acid classifications and physicochemical properties. In order to speed up our system, an automatic search method for the kernel parameter is used. The prediction performance of our method is evaluated on four datasets: Lei dataset, multi-localization dataset, SNL9 dataset and a new independent dataset. The overall accuracy of prediction for 6 localizations on Lei dataset is 75.2% and that for 9 localizations on SNL9 dataset is 72.1% in the leave-one-out cross validation, 71.7% for the multi-localization dataset and 69.8% for the new independent dataset, respectively. Comparisons with those existing methods show that our method performs better for both single-localization and multi-localization proteins and achieves more balanced sensitivities and specificities on large-size and small-size subcellular localizations. The overall accuracy improvements are 4.0% and 4.7% for single-localization proteins and 6.5% for multi-localization proteins. The reliability and stability of our classification model are further confirmed by permutation analysis. Conclusions It can be concluded that our method is effective and valuable for predicting protein subnuclear localizations. A web server has been designed to implement the proposed method. It is freely available at http://bioinformatics.awowshop.com/snlpr​ed_page.php.

The heterogeneous cluster ensemble method using hubness for clustering text documents

We propose a cluster ensemble method to map the corpus documents into the semantic space embedded in Wikipedia and group them using multiple types of feature space. A heterogeneous cluster ensemble is constructed with multiple types of relations i.e. document-term, document-concept and document-category. A final clustering solution is obtained by exploiting associations between document pairs and hubness of the documents. Empirical analysis with various real data sets reveals that the proposed meth-od outperforms state-of-the-art text clustering approaches.

The adventure of the refrain : composing with improvised music

This practice-based inquiry investigates the process of composing notated scores using improvised solos by saxophonists John Butcher and Anthony Braxton. To compose with these improvised sources, I developed a new method of analysis and through this method I developed new compositional techniques in applying these materials into a score. This method of analysis and composition utilizes the conceptual language of Gilles Deleuze and Felix Guattari found in A Thousand Plateaus. The conceptual language of Deleuze and Guattari, in particular the terms assemblage, refrain and deterritorialization are discussed in depth to give a context for the philosophical origins and also to explain how the language is used in reference to improvised music and the compositional process. The project seeks to elucidate the conceptual language through the creative practice and in turn for the creative practice to clarify the use of the conceptual terminology. The outcomes of the research resulted in four notated works being composed. Firstly, Gravity, for soloist and ensemble based on the improvisational language of John Butcher and secondly a series of 3 studies titled Transbraxton Studies for solo instruments based on the improvisational-compositional language of Anthony Braxton. The implications of this research include the application of the analysis method to a number of musical contexts including: to be used in the process of composing with improvised music; in the study of style and authorship in solo improvisation; as a way of analyzing group improvisation; in the analysis of textural music including electronic music; and in the analysis of music from different cultures—particularly cultures where improvisation and per formative aspects to the music are significant to the overall meaning of the work. The compositional technique that was developed has further applications in terms of an expressive method of composing with non-metered improvised materials and one that merges well with the transcription method developed of notating pitch and sounds to a timeline. It is hoped that this research can open further lines of enquiry into the application of the conceptual ideas of Deleuze and Guattari to the analysis of more forms of music.

Pulse: The London Symphony Orchestra Student Mobile Project

This report documents the key findings of a year-long collaborative research project focusing on the London Symphony Orchestra’s (LSO) development, implementation and testing of a mobile ticketing and information system. This ticketing system was developed in association with the LSO’s technical partners, Kodime Limited and in collaboration with the Aurora Orchestra.

PV string per-module maximum power point enabling converters

Many grid connected PV installations consist of a single series string of PV modules and a single DC-AC inverter. This efficiency of this topology can be enhanced with additional low power, low cost per panel converter modules. Most current flows directly in the series string which ensures high efficiency. However parallel Cúk or buck-boost DC-DC converters connected across each adjacent pair of modules now support any desired current difference between series connected PV modules. Each converter “shuffles” the desired difference in PV module currents between two modules and so on up the string. Spice simulations show that even with poor efficiency, these modules can make a significant improvement to the overall power which can be recovered from partially shaded PV strings.

Unsupervised temporal ensemble alignment for rapid annotation

This paper presents a novel framework for the unsupervised alignment of an ensemble of temporal sequences. This approach draws inspiration from the axiom that an ensemble of temporal signals stemming from the same source/class should have lower rank when "aligned" rather than "misaligned". Our approach shares similarities with recent state of the art methods for unsupervised images ensemble alignment (e.g. RASL) which breaks the problem into a set of image alignment problems (which have well known solutions i.e. the Lucas-Kanade algorithm). Similarly, we propose a strategy for decomposing the problem of temporal ensemble alignment into a similar set of independent sequence problems which we claim can be solved reliably through Dynamic Time Warping (DTW). We demonstrate the utility of our method using the Cohn-Kanade+ dataset, to align expression onset across multiple sequences, which allows us to automate the rapid discovery of event annotations.

Physical activity recognition from accelerometer data using a multi-scale ensemble method

Accurate and detailed measurement of an individual's physical activity is a key requirement for helping researchers understand the relationship between physical activity and health. Accelerometers have become the method of choice for measuring physical activity due to their small size, low cost, convenience and their ability to provide objective information about physical activity. However, interpreting accelerometer data once it has been collected can be challenging. In this work, we applied machine learning algorithms to the task of physical activity recognition from triaxial accelerometer data. We employed a simple but effective approach of dividing the accelerometer data into short non-overlapping windows, converting each window into a feature vector, and treating each feature vector as an i.i.d training instance for a supervised learning algorithm. In addition, we improved on this simple approach with a multi-scale ensemble method that did not need to commit to a single window size and was able to leverage the fact that physical activities produced time series with repetitive patterns and discriminative features for physical activity occurred at different temporal scales.

Steady state genuine multipartite entanglement in harmonic oscillator ensembles with a common environment

Steady state entanglement in ensembles of harmonic oscillators with a common squeezed reservoir is studied. Under certain conditions the ensemble features genuine multipartite entanglement in the steady state. Several analytic results regarding the bipartite and multipartite entanglement properties of the system are derived. We also discuss a possible experimental implementation which may exhibit steady state genuine multipartite entanglement.

The kite string tangle : blissed-out beats at the Brisbane Festival

Danny Harley has had a pretty big year with his one-man band project The Kite String Tangle. There have been festival appearances, both at home and abroad, a song in the top 20 of Triple J’s iconic Hottest 100 annual music poll and a sold-out tour to mark the release of his debut EP, which wrapped up this week with two shows in the Spiegeltent at the Brisbane Festival...

Text mining with semantic annotation : using enriched text representation for entity-oriented retrieval, semantic relation identification and text clustering

This project is a step forward in the study of text mining where enhanced text representation with semantic information plays a significant role. It develops effective methods of entity-oriented retrieval, semantic relation identification and text clustering utilizing semantically annotated data. These methods are based on enriched text representation generated by introducing semantic information extracted from Wikipedia into the input text data. The proposed methods are evaluated against several start-of-art benchmarking methods on real-life data-sets. In particular, this thesis improves the performance of entity-oriented retrieval, identifies different lexical forms for an entity relation and handles clustering documents with multiple feature spaces.

Robust clustering of multi-type relational data via a heterogeneous manifold ensemble

High-Order Co-Clustering (HOCC) methods have attracted high attention in recent years because of their ability to cluster multiple types of objects simultaneously using all available information. During the clustering process, HOCC methods exploit object co-occurrence information, i.e., inter-type relationships amongst different types of objects as well as object affinity information, i.e., intra-type relationships amongst the same types of objects. However, it is difficult to learn accurate intra-type relationships in the presence of noise and outliers. Existing HOCC methods consider the p nearest neighbours based on Euclidean distance for the intra-type relationships, which leads to incomplete and inaccurate intra-type relationships. In this paper, we propose a novel HOCC method that incorporates multiple subspace learning with a heterogeneous manifold ensemble to learn complete and accurate intra-type relationships. Multiple subspace learning reconstructs the similarity between any pair of objects that belong to the same subspace. The heterogeneous manifold ensemble is created based on two-types of intra-type relationships learnt using p-nearest-neighbour graph and multiple subspaces learning. Moreover, in order to make sure the robustness of clustering process, we introduce a sparse error matrix into matrix decomposition and develop a novel iterative algorithm. Empirical experiments show that the proposed method achieves improved results over the state-of-art HOCC methods for FScore and NMI.

Hybrid stiff-string-polynomial basis functions for vibration analysis of high speed rotating beams

A new finite element is developed for free vibration analysis of high speed rotating beams using basis functions which use a linear combination of the solution of the governing static differential equation of a stiff-string and a cubic polynomial. These new shape functions depend on rotation speed and element position along the beam and account for the centrifugal stiffening effect. The natural frequencies predicted by the proposed element are compared with an element with stiff-string, cubic polynomial and quintic polynomial shape functions. It is found that the new element exhibits superior convergence compared to the other basis functions.