International development: exploring the gap between organisations’ development policy and practice—a Southern perspective

International development policies inevitably encounter a conflict in their implementation, representing the gap between universal goals and grass-roots practice. The aim of this study was to explore and understand the significance of this gap, and to apply knowledge management principles as a lens to suggest bridging solutions. The research focuses on non-governmental organisations, which are a sub-section of the civil society. The study was unique as it took a Southern perspective—the views and experiences of policy-makers, practitioners and beneficiaries in Kenya form the basis of the empirical research. Consequently, the research and its findings provide some answers as to how systems of knowledge can be aligned with the core values of organisational and societal culture in order to positively influence international stability.     

Global Stereo Reconstruction under Second-Order Smoothness Priors

Second-order priors on the smoothness of 3D surfaces are a better model of typical scenes than first-order priors. However, stereo reconstruction using global inference algorithms, such as graph cuts, has not been able to incorporate second-order priors because the triple cliques needed to express them yield intractable (nonsubmodular) optimization problems. This paper shows that inference with triple cliques can be effectively performed. Our optimization strategy is a development of recent extensions to alpha-expansion, based on the ldquo QPBOrdquo algorithm. The strategy is to repeatedly merge proposal depth maps using a novel extension of QPBO. Proposal depth maps can come from any source, for example, frontoparallel planes as in alpha-expansion, or indeed any existing stereo algorithm, with arbitrary parameter settings.     

Singular perturbation theory for predicting extravasation of Brownian particles

Motivated by recent studies on tumor treatments using the drug delivery of nanoparticles, we provide a singular perturbation theory and perform Brownian dynamics simulations to quantify the extravasation rate of Brownian particles in a shear flow over a circular pore with a lumped mass transfer resistance. The analytic theory we present is an expansion in the limit of a vanishing Péclet number ( $P$ ), which is the ratio of convective fluxes to diffusive fluxes on the length scale of the pore. We state the concentration of particles near the pore and the extravasation rate (Sherwood number) to $O(P^{1/2})$ . This model improves upon previous studies because the results are valid for all values of the particle mass transfer coefficient across the pore, as modeled by the Damköhler number ( $\kappa $ ), which is the ratio of the reaction rate to the diffusive mass transfer rate at the boundary. Previous studies focused on the adsorption-dominated regime (i.e., $\kappa \rightarrow \infty $ ). Specifically, our work provides a theoretical basis and an interpolation-based approximate method for calculating the Sherwood number (a measure of the extravasation rate) for the case of finite resistance [ $\kappa \sim O(1)$ ] at small Péclet numbers, which are physiologically important in the extravasation of nanoparticles. We compare the predictions of our theory and an approximate method to Brownian dynamics simulations with reflection–reaction boundary conditions as modeled by $\kappa $ . They are found to agree well at small $P$ and for the $\kappa \ll 1$ and $\kappa \gg 1$ asymptotic limits representing the diffusion-dominated and adsorption-dominated regimes, respectively. Although this model neglects the finite size effects of the particles, it provides an important first step toward understanding the physics of extravasation in the tumor vasculature.     

Obese people who seek treatment have different characteristics than those who do not seek treatment.

A group of obese people who had not sought treatment; an obese group who had sought treatment in a professional, hospital-based program; and normal-weight controls (N?=?547) were compared in regard to level of psychopathology, binge eating, and negative emotional eating. Because the groups differed significantly on several demographic variables, 3 demographically matched groups were created and compared (n?=?177, 59 per group). In the matched subgroups, obese people who had sought treatment reported greater psychopathology and more binge eating than did normal-weight controls or those who had not sought treatment. Both obese groups (including those who had not sought treatment) endorsed more symptoms of distress, negative emotional eating, overeating, difficulty resisting temptation, and less exercise than did normal-weight controls. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The Problem of Degeneracy in Structure and Motion Recovery from Uncalibrated Image Sequences

The aim of this work is the recovery of 3D structure and camera projection matrices for each frame of an uncalibrated image sequence. In order to achieve this, correspondences are required throughout the sequence. A significant and successful mechanism for automatically establishing these correspondences is by the use of geometric constraints arising from scene rigidity. However, problems arise with such geometry guided matching if general viewpoint and general structure are assumed whilst frames in the sequence and/or scene structure do not conform to these assumptions. Such cases are termed degenerate.In this paper we describe two important cases of degeneracy and their effects on geometry guided matching. The cases are a motion degeneracy where the camera does not translate between frames, and a structure degeneracy where the viewed scene structure is planar. The effects include the loss of correspondences due to under or over fitting of geometric models estimated from image data, leading to the failure of the tracking method. These degeneracies are not a theoretical curiosity, but commonly occur in real sequences where models are statistically estimated from image points with measurement error.We investigate two strategies for tackling such degeneracies: the first uses a statistical model selection test to identify when degeneracies occur: the second uses multiple motion models to overcome the degeneracies. The strategies are evaluated on real sequences varying in motion, scene type, and length from 13 to 120 frames.     

PMBP: PatchMatch Belief Propagation for Correspondence Field Estimation

PatchMatch (PM) is a simple, yet very powerful and successful method for optimizing continuous labelling problems. The algorithm has two main ingredients: the update of the solution space by sampling and the use of the spatial neighbourhood to propagate samples. We show how these ingredients are related to steps in a specific form of belief propagation (BP) in the continuous space, called max-product particle BP (MP-PBP). However, MP-PBP has thus far been too slow to allow complex state spaces. In the case where all nodes share a common state space and the smoothness prior favours equal values, we show that unifying the two approaches yields a new algorithm, PMBP, which is more accurate than PM and orders of magnitude faster than MP-PBP. To illustrate the benefits of our PMBP method we have built a new stereo matching algorithm with unary terms which are borrowed from the recent PM Stereo work and novel realistic pairwise terms that provide smoothness. We have experimentally verified that our method is an improvement over state-of-the-art techniques at sub-pixel accuracy level.     

Design Role-Playing at the School of Architecture

A client-architect activity introduced into the second-year program at the School of Architecture, Washington University, St. Louis.     

Get out of MySpace!

To understand the student experience on social software, the research aims to explore the disruptive nature and opportunity of social networking for higher education. Taking four universities, the research: (1) identifies the distinction between the students’ current usage of social software; (2) reports on the students’ experience on opportunities and challenges of learning with social software; and (3) introduces principles as a guideline in using social software for learning. Quantitative research methods (web-based questionnaires) were incorporated to investigate the pattern of learners’ usage. Qualitative methods (student interviews) were adopted to clarify and further inform this relationship and their attitudes towards social software for learning. The results demonstrate a massive use of educational technology with distinct divide between the learning space and personal space. Student voices reveal that the central problem of such divide is due to the contrast perception and experience of ‘learning/studying and social life’. We argue that online learning and social personas may overlap but that learning needs to be designed so that it addresses the individual preferences to combine or separate the two domains. The paper concludes with a few principles of learning with social software grounded in students’ experience and Vygotsky’s paradigm.