Evaluating the use of a very large-scale presentation and collaboration framework

In this paper we describe the use and evaluation of CubIT, a multi-user, very large-scale presentation and collaboration framework. CubIT is installed at the Queensland University of Technology’s (QUT) Cube facility. The “Cube” is an interactive visualisation facility made up of five very large-scale interactive multi-panel wall displays, each consisting of up to twelve 55-inch multi-touch screens (48 screens in total) and massive projected display screens situated above the display panels. The paper outlines the unique design challenges, features, use and evaluation of CubIT. The system was built to make the Cube facility accessible to QUT’s academic and student population. CubIT enables users to easily upload and share their own media content, and allows multiple users to simultaneously interact with the Cube’s wall displays. The features of CubIT are implemented via three user interfaces, a multi-touch interface working on the wall displays, a mobile phone and tablet application and a web-based content management system. The evaluation reveals issues around the public use and functional scope of the system.

Random Gabor based templates for facial expression recognition in images with facial occlusion

Robust facial expression recognition (FER) under occluded face conditions is challenging. It requires robust algorithms of feature extraction and investigations into the effects of different types of occlusion on the recognition performance to gain insight. Previous FER studies in this area have been limited. They have spanned recovery strategies for loss of local texture information and testing limited to only a few types of occlusion and predominantly a matched train-test strategy. This paper proposes a robust approach that employs a Monte Carlo algorithm to extract a set of Gabor based part-face templates from gallery images and converts these templates into template match distance features. The resulting feature vectors are robust to occlusion because occluded parts are covered by some but not all of the random templates. The method is evaluated using facial images with occluded regions around the eyes and the mouth, randomly placed occlusion patches of different sizes, and near-realistic occlusion of eyes with clear and solid glasses. Both matched and mis-matched train and test strategies are adopted to analyze the effects of such occlusion. Overall recognition performance and the performance for each facial expression are investigated. Experimental results on the Cohn-Kanade and JAFFE databases demonstrate the high robustness and fast processing speed of our approach, and provide useful insight into the effects of occlusion on FER. The results on the parameter sensitivity demonstrate a certain level of robustness of the approach to changes in the orientation and scale of Gabor filters, the size of templates, and occlusions ratios. Performance comparisons with previous approaches show that the proposed method is more robust to occlusion with lower reductions in accuracy from occlusion of eyes or mouth.

Matching image sets via adaptive multi convex hull

Traditional nearest points methods use all the samples in an image set to construct a single convex or affine hull model for classification. However, strong artificial features and noisy data may be generated from combinations of training samples when significant intra-class variations and/or noise occur in the image set. Existing multi-model approaches extract local models by clustering each image set individually only once, with fixed clusters used for matching with various image sets. This may not be optimal for discrimination, as undesirable environmental conditions (eg. illumination and pose variations) may result in the two closest clusters representing different characteristics of an object (eg. frontal face being compared to non-frontal face). To address the above problem, we propose a novel approach to enhance nearest points based methods by integrating affine/convex hull classification with an adapted multi-model approach. We first extract multiple local convex hulls from a query image set via maximum margin clustering to diminish the artificial variations and constrain the noise in local convex hulls. We then propose adaptive reference clustering (ARC) to constrain the clustering of each gallery image set by forcing the clusters to have resemblance to the clusters in the query image set. By applying ARC, noisy clusters in the query set can be discarded. Experiments on Honda, MoBo and ETH-80 datasets show that the proposed method outperforms single model approaches and other recent techniques, such as Sparse Approximated Nearest Points, Mutual Subspace Method and Manifold Discriminant Analysis.

Multi-shot person re-identification via relational stein divergence

Person re-identification is particularly challenging due to significant appearance changes across separate camera views. In order to re-identify people, a representative human signature should effectively handle differences in illumination, pose and camera parameters. While general appearance-based methods are modelled in Euclidean spaces, it has been argued that some applications in image and video analysis are better modelled via non-Euclidean manifold geometry. To this end, recent approaches represent images as covariance matrices, and interpret such matrices as points on Riemannian manifolds. As direct classification on such manifolds can be difficult, in this paper we propose to represent each manifold point as a vector of similarities to class representers, via a recently introduced form of Bregman matrix divergence known as the Stein divergence. This is followed by using a discriminative mapping of similarity vectors for final classification. The use of similarity vectors is in contrast to the traditional approach of embedding manifolds into tangent spaces, which can suffer from representing the manifold structure inaccurately. Comparative evaluations on benchmark ETHZ and iLIDS datasets for the person re-identification task show that the proposed approach obtains better performance than recent techniques such as Histogram Plus Epitome, Partial Least Squares, and Symmetry-Driven Accumulation of Local Features.

Developing a multidimensional scale of customer-oriented deviance (COD)

Although frontline employees' bending of organizational rules and norms for customers is an important phenomenon, marketing scholars to date only broadly describe over-servicing behaviors and provide little distinction among deviant behavioral concepts. Drawing on research on pro-social and pro-customer behaviors and on studies of positive deviance, this paper develops and validates a multi-faceted, multi-dimensional construct term customer-oriented deviance. Results from two samples totaling 616 frontline employees (FLEs) in the retail and hospitality industries demonstrate that customer-oriented deviance is a four-dimensional construct with sound psychometric properties. Evidence from a test of a theoretical model of key antecedents establishes nomological validity with empathy/perspective-taking, risk-taking propensity, role conflict, and job autonomy as key predictors. Results show that the dimensions of customer-oriented deviance are distinct and have significant implications for theory and practice.

The validity and reliability of the Parent Fever Management Scale: A study from Palestine

Background Parental fever phobia and overuse of antipyretics to control fever is increasing. Little is known about childhood fever management among Arab parents. No scales to measure parents’ fever management practices in Palestine are available. Aims The aims of this study were to translate and examine the psychometric properties of the Arabic version of the Parent Fever Management Scale (PFMS). Methods A standard “forward–backward” procedure was used to translate PFMS into Arabic language. It was then validated on a convenience sample of 402 parents between July and October 2012. Descriptive statistics were used, and instrument reliability was assessed for internal consistency using Cronbach's alpha coefficient. Validity was confirmed using convergent and known group validation. Results Applying the recommended scoring method, the median (interquartile range) score of the PFMS was 26 (23-30). Acceptable internal consistency was found (Cronbach’s alpha = 0.733) and the test–retest reliability value was 0.92 (P < 0.001). The chi-squared (χ2) test showed a significant relationship between PFMS groups and frequent daily administration of antipyretic groups (χ2 = 52.86; P < 0.001). The PFMS sensitivity and specificity were 77.67% and 57.75%, respectively. The positive and negative predictive values were 67.89% and 32.11%, respectively. Conclusions The findings of this validation study indicate that the Arabic version of the PFMS is a reliable and valid measure which can be used as a useful tool for health professionals to identify parents’ fever management practices and thus provide targeted education to reduce the unnecessary burden of care they place on themselves when concerned for a febrile child.

STAC living. Promoting efficient multi-residential buildings in subtropical climates.

STAC is a mobile application (app) designed to promote the benefits of climate-aware urban development in Subtropical environments. Although, STAC is primarily tool for understanding climate efficient buildings in Brisbane, Australia, it also demonstrates how other exemplary buildings operate in other subtropical cities of the world. The STAC research and development team applied research undertaken by the Centre for Subtropical Design (Brisbane) to profile buildings past and present that have contributed to the creation of a vibrant society, a viable economy, a healthy environment, and an authentic sense of place. In collaboration with researchers from the field of Interaction Design, this knowledge and data was collated, processed and curated for presentation via a custom mobile application designed to distribute this important research for review and consideration on-location in local settings and for comparison across all other global subtropical regions and projects identified by this research. This collaboration adopted a Design-based Research (DBR) Methodology guided by the main tenets of research and design iteration and cross-discipline collaboration in real-world settings, resulting in the formulation of contextually-sensitive design principles, theories, and tools for design intervention. Combined with significant context review of available technology and data and subsequent case study analysis of exemplar design applications.

Supporting collaboration on very large-scale interactive wall surfaces

In this paper we describe CubIT, a multi-user presentation and collaboration system installed at the Queensland University of Technology’s (QUT) Cube facility. The ‘Cube’ is an interactive visualisation facility made up of five very large-scale interactive multi-panel wall displays, each consisting of up to twelve 55-inch multi-touch screens (48 screens in total) and massive projected display screens situated above the display panels. The paper outlines the unique design challenges, features, implementation and evaluation of CubIT. The system was built to make the Cube facility accessible to QUT’s academic and student population. CubIT enables users to easily upload and share their own media content, and allows multiple users to simultaneously interact with the Cube’s wall displays. The features of CubIT were implemented via three user interfaces, a multi-touch interface working on the wall displays, a mobile phone and tablet application and a web-based content management system. Each of these interfaces plays a different role and offers different interaction mechanisms. Together they support a wide range of collaborative features including multi-user shared workspaces, drag and drop upload and sharing between users, session management and dynamic state control between different parts of the system. The results of our evaluation study showed that CubIT was successfully used for a variety of tasks, and highlighted challenges with regards to user expectations regarding functionality as well as issues arising from public use.

An efficient and robust system for multi-person event detection in real world indoor surveillance scenes

Due to the popularity of security cameras in public places, it is of interest to design an intelligent system that can efficiently detect events automatically. This paper proposes a novel algorithm for multi-person event detection. To ensure greater than real-time performance, features are extracted directly from compressed MPEG video. A novel histogram-based feature descriptor that captures the angles between extracted particle trajectories is proposed, which allows us to capture motion patterns of multi-person events in the video. To alleviate the need for fine-grained annotation, we propose the use of Labelled Latent Dirichlet Allocation, a “weakly supervised” method that allows the use of coarse temporal annotations which are much simpler to obtain. This novel system is able to run at approximately ten times real-time, while preserving state-of-theart detection performance for multi-person events on a 100-hour real-world surveillance dataset (TRECVid SED).

Large-scale analysis of soccer matches using spatiotemporal tracking data

Although the collection of player and ball tracking data is fast becoming the norm in professional sports, large-scale mining of such spatiotemporal data has yet to surface. In this paper, given an entire season's worth of player and ball tracking data from a professional soccer league (approx 400,000,000 data points), we present a method which can conduct both individual player and team analysis. Due to the dynamic, continuous and multi-player nature of team sports like soccer, a major issue is aligning player positions over time. We present a "role-based" representation that dynamically updates each player's relative role at each frame and demonstrate how this captures the short-term context to enable both individual player and team analysis. We discover role directly from data by utilizing a minimum entropy data partitioning method and show how this can be used to accurately detect and visualize formations, as well as analyze individual player behavior.

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.

Large scale multiuser digital mind mapping tool : a study in usability

Project work can involve multiple people from varying disciplines coming together to solve problems as a group. Large scale interactive displays are presenting new opportunities to support such interactions with interactive and semantically enabled cooperative work tools such as intelligent mind maps. In this paper, we present a novel digital, touch-enabled mind-mapping tool as a first step towards achieving such a vision. This first prototype allows an evaluation of the benefits of a digital environment for a task that would otherwise be performed on paper or flat interactive surfaces. Observations and surveys of 12 participants in 3 groups allowed the formulation of several recommendations for further research into: new methods for capturing text input on touch screens; inclusion of complex structures; multi-user environments and how users make the shift from single- user applications; and how best to navigate large screen real estate in a touch-enabled, co-present multi-user setting.

Large Scale Simulations of Hippocampal-Neocortical Interactions in a Parallel Version of Genesis

A hippocampal-CA3 memory model was constructed with PGENESIS, a recently developed version of GENESIS that allows for distributed processing of a neural network simulation. A number of neural models of the human memory system have identified the CA3 region of the hippocampus as storing the declarative memory trace. However, computational models designed to assess the viability of the putative mechanisms of storage and retrieval have generally been too abstract to allow comparison with empirical data. Recent experimental evidence has shown that selective knock-out of NMDA receptors in the CA1 of mice leads to reduced stability of firing specificity in place cells. Here a similar reduction of stability of input specificity is demonstrated in a biologically plausible neural network model of the CA3 region, under conditions of Hebbian synaptic plasticity versus an absence of plasticity. The CA3 region is also commonly associated with seizure activity. Further simulations of the same model tested the response to continuously repeating versus randomized nonrepeating input patterns. Each paradigm delivered input of equal intensity and duration. Non-repeating input patterns elicited a greater pyramidal cell spike count. This suggests that repetitive versus non-repeating neocortical inpus has a quantitatively different effect on the hippocampus. This may be relevant to the production of independent epileptogenic zones and the process of encoding new memories.

Development of a pilot-scale bacterial fermentation for plasmid-based biopharmaceutical production using a stoichiometric medium

The recognition of the potential efficacy of plasmid DNA (pDNA) molecules as vectors in the treatment and prevention of emerging diseases has birthed the confidence to combat global pandemics. This is due to the close-to-zero safety concern associated with pDNA vectors compared to viral vectors in cell transfection and targeting. Considerable attention has been paid to the potential of pDNA vectors but comparatively less thought has been given to the practical challenges in producing large quantities to meet current rising demands. A pilot-scale fermentation scheme was developed by employing a stoichiometrically-designed growth medium whose exceptional plasmid yield performance was attested in a shake flask environment for pUC19 and pEGFP-N1 transformed into E. coliDH5α and E. coliJM109, respectively. Batch fermentation of E. coliDH5α-pUC19 employing the stoichiometric medium displayed a maximum plasmid volumetric and specific yield of 62.6 mg/L and 17.1 mg/g (mg plasmid/g dry cell weight), respectively. Fed-batch fermentation of E. coliDH5α-pUC19 on a glycerol substrate demonstrated one of the highest ever reported pilot-scale plasmid specific yield of 48.98 mg/g and a volumetric yield of 0.53 g/L. The attainment of high plasmid specific yields constitutes a decrease in plasmid manufacturing cost and enhances the effectiveness of downstream processes by reducing the proportion of intracellular impurities. The effect of step-rise temperature induction was also considered to maximize ColE1-origin plasmid replication.

Scale adaptive tracking using mean shift and efficient feature matching

The mean shift tracker has achieved great success in visual object tracking due to its efficiency being nonparametric. However, it is still difficult for the tracker to handle scale changes of the object. In this paper, we associate a scale adaptive approach with the mean shift tracker. Firstly, the target in the current frame is located by the mean shift tracker. Then, a feature point matching procedure is employed to get the matched pairs of the feature point between target regions in the current frame and the previous frame. We employ FAST-9 corner detector and HOG descriptor for the feature matching. Finally, with the acquired matched pairs of the feature point, the affine transformation between target regions in the two frames is solved to obtain the current scale of the target. Experimental results show that the proposed tracker gives satisfying results when the scale of the target changes, with a good performance of efficiency.