Probabilistic matching of image sets for video-based face recognition

We address the problem of face recognition on video by employing the recently proposed probabilistic linear discrimi-nant analysis (PLDA). The PLDA has been shown to be robust against pose and expression in image-based face recognition. In this research, the method is extended and applied to video where image set to image set matching is performed. We investigate two approaches of computing similarities between image sets using the PLDA: the closest pair approach and the holistic sets approach. To better model face appearances in video, we also propose the heteroscedastic version of the PLDA which learns the within-class covariance of each individual separately. Our experi-ments on the VidTIMIT and Honda datasets show that the combination of the heteroscedastic PLDA and the closest pair approach achieves the best performance.

The implementation of performance based planning in Queensland under the integrated planning act 1997 : an evaluation of perceptions and planning schemes

Performance based planning is a form of planning regulation that is not well understood and the theoretical advantages of this type of planning are rarely achieved in practice. Normatively, this type of regulation relies on performance standards that are quantifiable and technically based which are designed to manage the effects of development, where performance standards provide certainty in respect of the level of performance and the means of achievement is flexible. Few empirical studies have attempted to examine how performance based planning has been conceptualised and implemented in practice. Existing literature is predominately anecdotal and consultant based (Baker et al. 2006) and has not sought to quantitatively examine how land use has been managed or determine how context influences implementation. The Integrated Planning Act 1997 (IPA) operated as Queensland’s principal planning legislation between March 1998 and December 2009. The IPA prevented Local Governments from prohibiting development or use and the term zone was absent from the legislation. While the IPA did not use the term performance based planning, the system is widely considered to be performance based in practice (e.g. Baker et al. 2006; Steele 2009a, 2009b). However, the degree to which the IPA and the planning system in Queensland is performance based is debated (e.g. Yearbury 1998; England 2004). Four research questions guided the research framework using Queensland as the case study. The questions sought to: determine if there is a common understanding of performance based planning; identify how performance based planning was expressed under the IPA; understand how performance based planning was implemented in plans; and explore the experiences of participants in the planning system. The research developed a performance adoption spectrum. The spectrum describes how performance based planning is implemented, ranging between pure and hybrid interpretations. An ex-post evaluation of seventeen IPA plans sought to determine plan performativity within the conceptual spectrum. Land use was examined from the procedural dimension of performance (Assessment Tables) and the substantive dimension of performance (Codes). A documentary analysis and forty one interviews supplemented the research. The analytical framework considered how context influenced performance based planning, including whether: the location of the local government affected land use management techniques; temporal variation in implementation exists; plan-making guidelines affected implementation; different perceptions of the concept exist; this type of planning applies to a range of spatial scales. Outcomes were viewed as the medium for determining the acceptability of development in Queensland, a significant departure from pure approaches found in the United States. Interviews highlighted the absence of plan-making direction in the IPA, which contributed to the confusion about the intended direction of the planning system and the myth that the IPA would guarantee a performance based system. A hybridised form of performance based planning evolved in Queensland which was dependent on prescriptive land use zones and specification of land use type, with some local governments going to extreme lengths to discourage certain activities in a predetermined manner. Context had varying degrees of influence on plan-making methods. Decision-making was found to be inconsistent and the system created a range of unforeseen consequences including difficulties associated with land valuation, increased development speculation, and the role of planners in court was found to be less critical than in the previous planning system.

CAT-SLAM : probabilistic localisation and mapping using a continuous appearance-based trajectory

This paper describes a new system, dubbed Continuous Appearance-based Trajectory Simultaneous Localisation and Mapping (CAT-SLAM), which augments sequential appearance-based place recognition with local metric pose filtering to improve the frequency and reliability of appearance-based loop closure. As in other approaches to appearance-based mapping, loop closure is performed without calculating global feature geometry or performing 3D map construction. Loop-closure filtering uses a probabilistic distribution of possible loop closures along the robot’s previous trajectory, which is represented by a linked list of previously visited locations linked by odometric information. Sequential appearance-based place recognition and local metric pose filtering are evaluated simultaneously using a Rao–Blackwellised particle filter, which weights particles based on appearance matching over sequential frames and the similarity of robot motion along the trajectory. The particle filter explicitly models both the likelihood of revisiting previous locations and exploring new locations. A modified resampling scheme counters particle deprivation and allows loop-closure updates to be performed in constant time for a given environment. We compare the performance of CAT-SLAM with FAB-MAP (a state-of-the-art appearance-only SLAM algorithm) using multiple real-world datasets, demonstrating an increase in the number of correct loop closures detected by CAT-SLAM.

Weighted LDA techniques for I-vector based speaker verification

This paper introduces the Weighted Linear Discriminant Analysis (WLDA) technique, based upon the weighted pairwise Fisher criterion, for the purposes of improving i-vector speaker verification in the presence of high intersession variability. By taking advantage of the speaker discriminative information that is available in the distances between pairs of speakers clustered in the development i-vector space, the WLDA technique is shown to provide an improvement in speaker verification performance over traditional Linear Discriminant Analysis (LDA) approaches. A similar approach is also taken to extend the recently developed Source Normalised LDA (SNLDA) into Weighted SNLDA (WSNLDA) which, similarly, shows an improvement in speaker verification performance in both matched and mismatched enrolment/verification conditions. Based upon the results presented within this paper using the NIST 2008 Speaker Recognition Evaluation dataset, we believe that both WLDA and WSNLDA are viable as replacement techniques to improve the performance of LDA and SNLDA-based i-vector speaker verification.

Approaches to learning information literacy : a phenomenographic study

This paper reports on an empirical study that explores the ways students approach learning to find and use information. Based on interviews with 15 education students in an Australian university, this study uses phenomenography as its methodological and theoretical basis. The study reveals that students use three main strategies for learning information literacy: 1) learning by doing; 2) learning by trial and error; and 3) learning by interacting with other people. Understanding the different ways that students approach learning information literacy will assist librarians and faculty to design and provide more effective information literacy education.

PLDA based speaker recognition on short utterances

This paper investigates the effects of limited speech data in the context of speaker verification using a probabilistic linear discriminant analysis (PLDA) approach. Being able to reduce the length of required speech data is important to the development of automatic speaker verification system in real world applications. When sufficient speech is available, previous research has shown that heavy-tailed PLDA (HTPLDA) modeling of speakers in the i-vector space provides state-of-the-art performance, however, the robustness of HTPLDA to the limited speech resources in development, enrolment and verification is an important issue that has not yet been investigated. In this paper, we analyze the speaker verification performance with regards to the duration of utterances used for both speaker evaluation (enrolment and verification) and score normalization and PLDA modeling during development. Two different approaches to total-variability representation are analyzed within the PLDA approach to show improved performance in short-utterance mismatched evaluation conditions and conditions for which insufficient speech resources are available for adequate system development. The results presented within this paper using the NIST 2008 Speaker Recognition Evaluation dataset suggest that the HTPLDA system can continue to achieve better performance than Gaussian PLDA (GPLDA) as evaluation utterance lengths are decreased. We also highlight the importance of matching durations for score normalization and PLDA modeling to the expected evaluation conditions. Finally, we found that a pooled total-variability approach to PLDA modeling can achieve better performance than the traditional concatenated total-variability approach for short utterances in mismatched evaluation conditions and conditions for which insufficient speech resources are available for adequate system development.

PLDA based speaker verification with weighted LDA techniques

This paper investigates the use of the dimensionality-reduction techniques weighted linear discriminant analysis (WLDA), and weighted median fisher discriminant analysis (WMFD), before probabilistic linear discriminant analysis (PLDA) modeling for the purpose of improving speaker verification performance in the presence of high inter-session variability. Recently it was shown that WLDA techniques can provide improvement over traditional linear discriminant analysis (LDA) for channel compensation in i-vector based speaker verification systems. We show in this paper that the speaker discriminative information that is available in the distance between pair of speakers clustered in the development i-vector space can also be exploited in heavy-tailed PLDA modeling by using the weighted discriminant approaches prior to PLDA modeling. Based upon the results presented within this paper using the NIST 2008 Speaker Recognition Evaluation dataset, we believe that WLDA and WMFD projections before PLDA modeling can provide an improved approach when compared to uncompensated PLDA modeling for i-vector based speaker verification systems.

Relative entropy rate based model selection for linear hybrid system filters of uncertain nonlinear systems

Hybrid system representations have been exploited in a number of challenging modelling situations, including situations where the original nonlinear dynamics are too complex (or too imprecisely known) to be directly filtered. Unfortunately, the question of how to best design suitable hybrid system models has not yet been fully addressed, particularly in the situations involving model uncertainty. This paper proposes a novel joint state-measurement relative entropy rate based approach for design of hybrid system filters in the presence of (parameterised) model uncertainty. We also present a design approach suitable for suboptimal hybrid system filters. The benefits of our proposed approaches are illustrated through design examples and simulation studies.

Challenges in mass spectrometry-based quantification of bioactive peptides: A case study exploring the neuropeptide Y family

The study of biologically active peptides is critical to the understanding of physiological pathways, especially those involved in the development of disease. Historically, the measurement of biologically active endogenous peptides has been undertaken by radioimmunoassay, a highly sensitive and robust technique that permits the detection of physiological concentrations in different biofluid and tissue extracts. Over recent years, a range of mass spectrometric approaches have been applied to peptide quantification with limited degrees of success. Neuropeptide Y (NPY), peptide YY (PYY), and pancreatic polypeptide (PP) belong to the NPY family exhibiting regulatory effects on appetite and feeding behavior. The physiological significance of these peptides depends on their molecular forms and in vivo concentrations systemically and at local sites within tissues. In this report, we describe an approach for quantification of individual peptides within mixtures using high-performance liquid chromatography electrospray ionization tandem mass spectrometry analysis of the NPY family peptides. Aspects of quantification including sample preparation, the use of matrix-matched calibration curves, and internal standards will be discussed. This method for the simultaneous determination of NPY, PYY, and PP was accurate and reproducible but lacks the sensitivity required for measurement of their endogenous concentration in plasma. The advantages of mass spectrometric quantification will be discussed alongside the current obstacles and challenges. © 2012 Wiley Periodicals, Inc. Biopolymers (Pept Sci) 98: 357–366, 2012.

Experimental comparison of odometry approaches

Odometry is an important input to robot navigation systems, and we are interested in the performance of vision-only techniques. In this paper we experimentally evaluate and compare the performance of wheel odometry, monocular feature-based visual odometry, monocular patch-based visual odometry, and a technique that fuses wheel odometry and visual odometry, on a mobile robot operating in a typical indoor environment.

Polycaprolactone-based scaffold plus recombinant human bone morphogenetic protein (rhBMP-2) in an ovine model of anterior spinal fusion

Synthetic scaffolds combined with growth factors have the potential to replace allograft or autograft as a graft material for spinal interbody fusion. Such tissue engineering approaches may be useful in Adolescent Idiopathic Scoliosis (AIS) surgery, however there are no studies to date examining the use of such biodegradable implants in combination with biologics in a thoracic spine model. This in vivo study examines the use of biodegradable polycaprolactone (PCL) based scaffolds with rhBMP-2 as a bone graft substitute in a sheep thoracic fusion model, where an anterior approach is used to simulate minimally invasive surgical deformity correction in the setting of AIS.

A least squares based finite volume method for the Cahn-Hilliard and Cahn-Hilliard-reaction equations

A vertex-centred finite volume method (FVM) for the Cahn-Hilliard (CH) and recently proposed Cahn-Hilliard-reaction (CHR) equations is presented. Information at control volume faces is computed using a high-order least-squares approach based on Taylor series approximations. This least-squares problem explicitly includes the variational boundary condition (VBC) that ensures that the discrete equations satisfy all of the boundary conditions. We use this approach to solve the CH and CHR equations in one and two dimensions and show that our scheme satisfies the VBC to at least second order. For the CH equation we show evidence of conservative, gradient stable solutions, however for the CHR equation, strict gradient-stability is more challenging to achieve.

A penalty-based grouping genetic algorithm for multiple composite SaaS components clustering in Cloud

Software as a Service (SaaS) in Cloud is getting more and more significant among software users and providers recently. A SaaS that is delivered as composite application has many benefits including reduced delivery costs, flexible offers of the SaaS functions and decreased subscription cost for users. However, this approach has introduced a new problem in managing the resources allocated to the composite SaaS. The resource allocation that has been done at the initial stage may be overloaded or wasted due to the dynamic environment of a Cloud. A typical data center resource management usually triggers a placement reconfiguration for the SaaS in order to maintain its performance as well as to minimize the resource used. Existing approaches for this problem often ignore the underlying dependencies between SaaS components. In addition, the reconfiguration also has to comply with SaaS constraints in terms of its resource requirements, placement requirement as well as its SLA. To tackle the problem, this paper proposes a penalty-based Grouping Genetic Algorithm for multiple composite SaaS components clustering in Cloud. The main objective is to minimize the resource used by the SaaS by clustering its component without violating any constraint. Experimental results demonstrate the feasibility and the scalability of the proposed algorithm.

Investigating the deployment of project management : a new perspective based on the concept of certification

Purpose – The purpose of this paper is to provide a summary description of the doctoral thesis investigating the field of project management (PM) deployment. Researchers will be informed of the current contributions within this topic and of the possible further investigations and researches. The decision makers and practitioners will be aware of a set of tools addressing the PM deployment with new perspectives. Design/methodology/approach – Research undertaken with the thesis is based on quantitative methods using time series statistics (time distance analysis) and comparative and correlation analysis aimed to better define and understand the PM deployment within and between countries or groups. Findings – The results suggest a project management deployment index (PMDI) to objectively measure the PM deployment based on the concept of certification. A proposed framework to empirically benchmark the PM deployment between countries by integrating the PMDI time series with the two dimensional comparative analysis of Sicherl. The correlation analysis within Hoftsede cultural framework shows the impact of the national culture dimensions on the PM deployment. The forecasting model shows a general continual growth trend of the PM deployment, with continual increase in the time distance between the countries. Research limitations/implications – The PM researchers are offered an empirical quantification on which they can construct further investigations and understanding of this phenomenon. The number of possible units that can be studied offers wide possibilities to replicate the thesis work. New researches can be undertaken to investigate further the contribution of other social or economical indicators, or to refine and enrich the definition of the PMDI indicator. Practical implications – These results have important implications on the PM deployment approaches. The PMDI measurements and time series comparisons facilitate considerably the measurement and benchmarking between the units (e.g. countries) and against targets, while the readiness setting of the studied unit (in terms of development and cultural levels) impacts the PM deployment within this country. Originality/value – This paper provides a summary of cutting-edge research work in the studied field of PM deployment and a link to the published works that researchers can use to help them understand the thesis research as well as how it can be extended.

Development of an innovative technology-based youth passenger safety program : an evidence-based approach

Young drivers are overrepresented in motor vehicle crash rates, and their risk increases when carrying similar aged passengers. Graduated Driver Licensing strategies have demonstrated effectiveness in reducing fatalities among young drivers, however complementary approaches may further reduce crash rates. Previous studies conducted by the researchers have shown that there is considerable potential for a passenger focus in youth road safety interventions, particularly involving the encouragement of young passengers to intervene in their peers’ risky driving (Buckley, Chapman, Sheehan & Davidson, 2012). Additionally, this research has shown that technology-based applications may be a promising means of delivering passenger safety messages, particularly as young people are increasingly accessing web-based and mobile technologies. This research describes the participatory design process undertaken to develop a web-based road safety program, and involves feasibility testing of storyboards for a youth passenger safety application. Storyboards and framework web-based materials were initially developed for a passenger safety program, using the results of previous studies involving online and school-based surveys with young people. Focus groups were then conducted with 8 school staff and 30 senior school students at one public high school in the Australian Capital Territory. Young people were asked about the situations in which passengers may feel unsafe and potential strategies for intervening in their peers’ risky driving. Students were also shown the storyboards and framework web-based material and were asked to comment on design and content issues. Teachers were also shown the material and asked about their perceptions of program design and feasibility. The focus group data will be used as part of the participatory design process, in further developing the passenger safety program. This research describes an evidence-based approach to the development of a web-based application for youth passenger safety. The findings of this research and resulting technology will have important implications for the road safety education of senior high school students.