Wide-baseline keypoint detection and matching with wide-angle images for vision based localisation

This thesis addresses the problem of detecting and describing the same scene points in different wide-angle images taken by the same camera at different viewpoints. This is a core competency of many vision-based localisation tasks including visual odometry and visual place recognition. Wide-angle cameras have a large field of view that can exceed a full hemisphere, and the images they produce contain severe radial distortion. When compared to traditional narrow field of view perspective cameras, more accurate estimates of camera egomotion can be found using the images obtained with wide-angle cameras. The ability to accurately estimate camera egomotion is a fundamental primitive of visual odometry, and this is one of the reasons for the increased popularity in the use of wide-angle cameras for this task. Their large field of view also enables them to capture images of the same regions in a scene taken at very different viewpoints, and this makes them suited for visual place recognition. However, the ability to estimate the camera egomotion and recognise the same scene in two different images is dependent on the ability to reliably detect and describe the same scene points, or ‘keypoints’, in the images. Most algorithms used for this purpose are designed almost exclusively for perspective images. Applying algorithms designed for perspective images directly to wide-angle images is problematic as no account is made for the image distortion. The primary contribution of this thesis is the development of two novel keypoint detectors, and a method of keypoint description, designed for wide-angle images. Both reformulate the Scale- Invariant Feature Transform (SIFT) as an image processing operation on the sphere. As the image captured by any central projection wide-angle camera can be mapped to the sphere, applying these variants to an image on the sphere enables keypoints to be detected in a manner that is invariant to image distortion. Each of the variants is required to find the scale-space representation of an image on the sphere, and they differ in the approaches they used to do this. Extensive experiments using real and synthetically generated wide-angle images are used to validate the two new keypoint detectors and the method of keypoint description. The best of these two new keypoint detectors is applied to vision based localisation tasks including visual odometry and visual place recognition using outdoor wide-angle image sequences. As part of this work, the effect of keypoint coordinate selection on the accuracy of egomotion estimates using the Direct Linear Transform (DLT) is investigated, and a simple weighting scheme is proposed which attempts to account for the uncertainty of keypoint positions during detection. A word reliability metric is also developed for use within a visual ‘bag of words’ approach to place recognition.

Contrast enhancement of EPID images via difference imaging : a feasibility study

In this study, the feasibility of difference imaging for improving the contrast of electronic portal imaging device (EPID) images is investigated. The difference imaging technique consists of the acquisition of two EPID images (with and without the placement of an additional layer of attenuating medium on the surface of the EPID)and the subtraction of one of these images from the other. The resulting difference image shows improved contrast, compared to a standard EPID image, since it is generated by lower-energy photons. Results of this study show that, ¯rstly, this method can produce images exhibiting greater contrast than is seen in standard megavoltage EPID images and that, secondly, the optimal thickness of attenuating material for producing a maximum contrast enhancement may vary with phantom thickness and composition. Further studies of the possibilities and limitations of the di®erence imaging technique, and the physics behind it, are therefore recommended.

A vision of the past : The Mill Albion

This paper examines The Mill Albion community history project, a diverse, multi-layered public history/art program that captures the social heritage of The Albion Flour Mill, as told through images produced as part of a research consultancy undertaken by QUT for FKP Property Group. The Albion Flour Mill was built in 1930 and continued operations for more than 72 years. After ceasing operation in 2005 the site was left to deteriorate. The FKP Property Group purchased the land to undertake a new urban redevelopment project. This paper reflects on the project and showcases some of the culturally creative ways this community’s history was told, using images.

Focus-score weighted super-resolution for uncooperative iris recognition at a distance and on the move

Uncooperative iris identification systems at a distance and on the move often suffer from poor resolution and poor focus of the captured iris images. The lack of pixel resolution and well-focused images significantly degrades the iris recognition performance. This paper proposes a new approach to incorporate the focus score into a reconstruction-based super-resolution process to generate a high resolution iris image from a low resolution and focus inconsistent video sequence of an eye. A reconstruction-based technique, which can incorporate middle and high frequency components from multiple low resolution frames into one desired super-resolved frame without introducing false high frequency components, is used. A new focus assessment approach is proposed for uncooperative iris at a distance and on the move to improve performance for variations in lighting, size and occlusion. A novel fusion scheme is then proposed to incorporate the proposed focus score into the super-resolution process. The experiments conducted on the The Multiple Biometric Grand Challenge portal database shows that our proposed approach achieves an EER of 2.1%, outperforming the existing state-of-the-art averaging signal-level fusion approach by 19.2% and the robust mean super-resolution approach by 8.7%.

Craft, process and art : teaching and learning music composition in higher education

This paper explores models of teaching and learning music composition in higher education. It analyses the pedagogical approaches apparent in the literature on teaching and learning composition in schools and universities, and introduces a teaching model as: learning from the masters; mastery of techniques; exploring ideas; and developing voice. It then presents a learning model developed from a qualitative study into students’ experiences of learning composition at university as: craft, process and art. The relationship between the students’ experiences and the pedagogical model is examined. Finally, the implications for composition curricula in higher education are presented.

PIPP #2 : Pedestrian Interaction Patch Project

Presented as part of the Sampled Festival at Sadlers Wells UK in January 2009, PIPP #2 continues the exploration of the first installation (PIPP #1 Leeds) and asks audiences to connect with an interactive work presented in the foyer of a major dance festival. Literally choreographing their own dances on the walls of the venue, pedestrians re-connect with the architectural surrounds generating unique memories of self.

The Silk Road project

The Silk Road Project was a practice-based research project investigating the potential of motion capture technology to inform perceptions of embodiment in dance performance. The project created a multi-disciplinary collaborative performance event using dance performance and real-time motion capture at Deakin University’s Deakin Motion Lab. Performances at Deakin University, December 2007.

Colonising Shakespeare? : agency and authority in Gregory Rogers's Shakespearean picture books

“You need to be able to tell stories. Illustration is a literature, not a pure fine art. It’s the fine art of writing with pictures.” – Gregory Rogers. This paper reads two recent wordless picture books by Australian illustrator Gregory Rogers in order to consider how “Shakespeare” is produced as a complex object of consumption for the implied child reader: The Boy, The Bear, The Baron, The Bard (2004) and Midsummer Knight (2006). In these books other worlds are constructed via time-travel and travel to a fantasy world, and clearly presume reader competence in narrative temporality and structure, and cultural literacy (particularly in reference to Elizabethan London and William Shakespeare), even as they challenge normative concepts via use of the fantastic. Exploring both narrative sequences and individual images reveals a tension in the books between past and present, and real and imagined. Where children’s texts tend to privilege Shakespeare, the man and his works, as inherently valuable, Rogers’s work complicates any sense of cultural value. Even as these picture books depend on a lexicon of Shakespearean images for meaning and coherence, they represent William Shakespeare as both an enemy to children (The Boy), and a national traitor (Midsummer). The protagonists, a boy in the first book and the bear he rescues in the second, effect political change by defeating Shakespeare. However, where these texts might seem to be activating a postcolonial cultural critique, this is complicated both by presumed readerly competence in authorized cultural discourses and by repeated affirmation of monarchies as ideal political systems. Power, then, in these picture books is at once rewarded and withheld, in a dialectic of (possibly postcolonial) agency, and (arguably colonial) subjection, even as they challenge dominant valuations of “Shakespeare” they do not challenge understandings of the “Child”.

V2V/V2I augmented maps : state-of-the-art and contribution to real-time crash risk assessment

A number of advanced driver assistance systems (ADAS) are currently being released on the market, providing safety functions to the drivers such as collision avoidance, adaptive cruise control or enhanced night-vision. These systems however are inherently limited by their sensory range: they cannot gather information from outside this range, also called their “perceptive horizon”. Cooperative systems are a developing research avenue that aims at providing extended safety and comfort functionalities by introducing vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) wireless communications to the road actors. This paper presents the problematic of cooperative systems, their advantages and contributions to road safety and exposes some limitations related to market penetration, sensors accuracy and communications scalability. It explains the issues of how to implement extended perception, a central contribution of cooperative systems. The initial steps of an evaluation of data fusion architectures for extended perception are exposed.

Customers behaving badly : a state of the art review, research agenda and implications for practitioners

The purpose of this paper is to highlight important issues in the study of dysfunctional customer behavior and to provide a research agenda to inspire, guide, and enthuse. Through a critical evaluation of existing research, the aim is to highlight key issues and to present potentially worthy avenues for future study.

Lumia : art/light/motion

Lumia: art/light/motion is an exciting new media exhibition presented by State Library of Queensland in partnership with Queensland-based Kuuki collective artists Priscilla Bracks and Gavin Sade. The exhibition explored contemporary life and encourages thought about the future through an extraordinary collection of hand-crafted and interactive electronic creatures and installations. The beautifully crafted new media artworks in Lumia: art/light/motion combine the bespoke with art and technology to create strange but intriguing objects. Lumia invited audiences to play, learn and then ponder the way we live and the environmental and social implications of our choices.

Uncertainties that matter : risk, response-ability, ethics and the moment of exchange in Live Art

This article examines the moment of exchange between artist, audience and culture in Live Art. Drawing on historical and contemporary examples, including examples from the Exist in 08 Live Art Event in Brisbane, Australia, in October 2008, it argues that Live Art - be it body art, activist art, site-specific performance, or other sorts of performative intervention in the public sphere - is characterised by a common set of claims about activating audiences, asking them to reflect on cultural norms challenged in the work. Live Art presents risky actions, in a context that blurs the boundaries between art and reality, to position audients as ‘witnesses’ who are personally implicated in, and responsible for, the actions unfolding before them. This article problematises assumptions about the way the uncertainties embedded in the Live Art encounter contribute to its deconstructive agenda. It uses the ethical theory of Emmanuel Levinas, Hans-Thies Lehmann and Dwight Conquergood to examine the mechanics of reductive, culturally-recuperative readings that can limit the efficacy of the Live Art encounter. It argues that, though ‘witnessing’ in Live Art depends on a relation to the real - real people, taking real risks, in real places - if it fails to foreground theatrical frame it is difficult for audients to develop the dual consciousness of the content, and their complicity in that content, that is the starting point for reflexivity, and response-ability, in the ethical encounter.

(Dia)logics of difference : disability, performance & spectatorship

In this paper I examine the way artists with disabilities use performances in public spaces to encourage people to reflect on their own contribution to the construction of publics, or counter-publics, during and after the moment of encounter. I focus on Liz Crow’s Resistance on the Plinth. This is the title Crow gave her performance on the Forth Plinth in Trafalgar Square as part of Antony Gormley’s One and Other project in 2009. Described as a public art project presenting a portrait of the U.K., Gormley’s One and Other gave 2400 people selected at random via a lottery the chance to do whatever they chose for an hour on the vacant Forth Plinth in Trafalgar Square. In her hour, Crow chose to present herself in a Nazi uniform, in her wheelchair. In this paper, I discuss how Crow’s performance used a counterposition of images – the Nazi uniform, associated with eugenics and a desire to eliminate people who do not accord with the Arayan ‘norm’, counterposed with her wheelchair – to encourage passersby to “stop, look, think.” I examine how Crow used this counterposition to prevent passersby from attributing a single, stable, monologic meaning to the image – forestalling the risk that passersby would read the image as a Nazi on the Plinth – and instead draw passersby into a dialogue in which the impossibility of reconciling the contradictory images, ideologies and cultural logics Crow embodied encouraged people to continue thinking and talking about these cultural logics during and after the encounter.

Stripping Sight Bare : Alison Jones’ ‘Portraits by Proxy’

In this paper I analyse UK artist Alison Jones’ sonic interventions Portrait of the Artist by Proxy (2008), Voyeurism by Proxy (2008) and Art, Lies and Audio Tapes (2009). In Portrait of the Artist by Proxy, Jones – who, due to deteriorating vision, has not seen her reflection in a mirror in years – asks and trusts participants to audio-describe her own image back to her. In Voyeurism by Proxy, Jones asks participants to audio-describe erotic drawings by Gustav Klimt. In Art, Lies and Audio Tapes, Jones asks participants to audio-describe other artworks, such as W.F. Yeames’ And When Did You Last see Your Father?. In these portraits by proxy, Jones opens her image, and other images, to interpretation. In doing so, Jones draws attention to the way sight is privileged as a mode of access to fixed, fundamental truths in Western culture – a mode assumed to be untainted by filters that skew perception of the object. “In a culture where vision is by far the dominant sense,” Jones says, “and as a visual artist with a visual impairment, I am reliant on audio-description …Inevitably, there are limitations imposed by language, time and the interpreter’s background knowledge of the subject viewed, as well as their personal bias of what is deemed important to impart in their description” . In these works, Jones strips these background knowledges, biases and assumptions bare. She reveals different perceptions, as well as tendencies or censor, edit or exaggerate descriptions. In this paper, I investigate how, by revealing unconscious biases, Jones’ works renders herself and her participants vulnerable to a change of perception. I also examine how Jones’ later editing of the audio-descriptions allows her to show the instabilities of sight, and, in Portrait of the Artist by Proxy, to reclaim authorship of her own image.

Digital processing of diffusion-tensor images of avascular tissues

Diffusion is the process that leads to the mixing of substances as a result of spontaneous and random thermal motion of individual atoms and molecules. It was first detected by the English botanist Robert Brown in 1827, and the phenomenon became known as ‘Brownian motion’. More specifically, the motion observed by Brown was translational diffusion – thermal motion resulting in random variations of the position of a molecule. This type of motion was given a correct theoretical interpretation in 1905 by Albert Einstein, who derived the relationship between temperature, the viscosity of the medium, the size of the diffusing molecule, and its diffusion coefficient. It is translational diffusion that is indirectly observed in MR diffusion-tensor imaging (DTI). The relationship obtained by Einstein provides the physical basis for using translational diffusion to probe the microscopic environment surrounding the molecule.