Unsupervised online learning of condition-invariant images for place recognition

This paper presents an online, unsupervised training algorithm enabling vision-based place recognition across a wide range of changing environmental conditions such as those caused by weather, seasons, and day-night cycles. The technique applies principal component analysis to distinguish between aspects of a location’s appearance that are condition-dependent and those that are condition-invariant. Removing the dimensions associated with environmental conditions produces condition-invariant images that can be used by appearance-based place recognition methods. This approach has a unique benefit – it requires training images from only one type of environmental condition, unlike existing data-driven methods that require training images with labelled frame correspondences from two or more environmental conditions. The method is applied to two benchmark variable condition datasets. Performance is equivalent or superior to the current state of the art despite the lesser training requirements, and is demonstrated to generalise to previously unseen locations.

Convolutional Neural Network-based place recognition

Recently Convolutional Neural Networks (CNNs) have been shown to achieve state-of-the-art performance on various classification tasks. In this paper, we present for the first time a place recognition technique based on CNN models, by combining the powerful features learnt by CNNs with a spatial and sequential filter. Applying the system to a 70 km benchmark place recognition dataset we achieve a 75% increase in recall at 100% precision, significantly outperforming all previous state of the art techniques. We also conduct a comprehensive performance comparison of the utility of features from all 21 layers for place recognition, both for the benchmark dataset and for a second dataset with more significant viewpoint changes.

Detecting rare events using Kullback-Leibler divergence

One main challenge in developing a system for visual surveillance event detection is the annotation of target events in the training data. By making use of the assumption that events with security interest are often rare compared to regular behaviours, this paper presents a novel approach by using Kullback-Leibler (KL) divergence for rare event detection in a weakly supervised learning setting, where only clip-level annotation is available. It will be shown that this approach outperforms state-of-the-art methods on a popular real-world dataset, while preserving real time performance.

Automated sensory data alignment for environmental and epidermal change monitoring

In this paper we present research adapting a state of the art condition-invariant robotic place recognition algorithm to the role of automated inter- and intra-image alignment of sensor observations of environmental and skin change over time. The approach involves inverting the typical criteria placed upon navigation algorithms in robotics; we exploit rather than attempt to fix the limited camera viewpoint invariance of such algorithms, showing that approximate viewpoint repetition is realistic in a wide range of environments and medical applications. We demonstrate the algorithms automatically aligning challenging visual data from a range of real-world applications: ecological monitoring of environmental change, aerial observation of natural disasters including flooding, tsunamis and bushfires and tracking wound recovery and sun damage over time and present a prototype active guidance system for enforcing viewpoint repetition. We hope to provide an interesting case study for how traditional research criteria in robotics can be inverted to provide useful outcomes in applied situations.

Perceiving the interactive body in dance: Enhancing kinesthetic empathy through art objects

Through a consideration of audience experience of embodiment in contemporary dance performance, this project used kinesthetic empathy as a theoretical construct to inform choreographic decision-making. The research outcome challenged the traditional performer/audience relationship through an interactive dance performance work entitled Planets. This acted as a platform that allowed both audience and performer to collaboratively listen to, process and form movement in a shared kinesthetic state. This connection was enabled through the distribution of interactive art objects, which responded to the shifting proximity between performer and audience. The performance was thus experienced through following a shared goal as instigated by the interactive technology. Through practice-led research, knowledge from kinesthetic empathy, embodied cognition and the mirror neuron system were used to develop the project’s aim in encouraging interactive audiences to engage in movement. This aim influenced studio explorations of movement through an enquiry into the kinesthetic self in dance. Investigations used movement quality, tension, mobility and acceleration to access a familiar movement vocabulary appropriate for a broad interactive audience. This informed the role of the researcher as performer. Planets was developed as a collaborative project between Michael Smith and interactive visual designer Andy Bates and performed over three nights at the Ars Electronica Festival 2014 in Linz, Austria. Supported by documented footage from Planets and audience responses to the performances, this paper draws together the theoretical underpinnings behind the development of the work and includes the experiential perspective of the performer.

Shifting intimacies

Our world is literally and figuratively turning to ‘dust’. This work acknowledges decay and renewal and the transitional, cyclical natures of interrelated ecologies. It also suggests advanced levels of degradation potentially beyond reparation. Dust exists both on and beneath the border of our unaided vision. Dust particles are predominantly forms of disintegrating solids that often become the substance or catalyst of future forms. Like many tiny forms, dust is an often unnoticed residue with ‘planet-size consequences’. (Hanna Holmes 2001) The image depicts an ethereal, backlit body, continually circling and morphing, apparently floating, suggesting endless cycles of birth, life and death and inviting differing states of meditation, exploration, stillness and play. This never ending video work is taken from a large-scale interactive/media artwork created during a six-month research residency in England at the Institute of Contemporary Art London and at Vincent Dance Theatre Sheffield in 2006. It was originally presented on a raised floor screen made of pure white sand at the ICA in London (see). The project involved developing new interaction, engagement and image making strategies for media arts practice, drawing on the application of both kinetic and proprioceptive dance/performance knowledges. The work was further informed by ecological network theory that assesses the systemic implications of private and public actions within bounded systems. The creative methodology was primarily practice-led which fomented the particular qualities of imagery, generated through cross-fertilising embodied knowledge of Dance and Media Arts. This was achieved through extensive workshopping undertaken in theatres, working ‘on the floor’ live, with dancers, props, sound and projection. And eventually of course, all this dust must settle. (Holmes 2001, from Dust Jacket) Holmes, H. 2001, The Secret Life of Dust: From the Cosmos to the Kitchen Counter, the Big Consequences of Little Things, p.3

Knowledge gain or system benefit in environmental decision making?

The quality of environmental decisions are gauged according to the management objectives of a conservation project. Management objectives are generally about maximising some quantifiable measure of system benefit, for instance population growth rate. They can also be defined in terms of learning about the system in question, in such a case actions would be chosen that maximise knowledge gain, for instance in experimental management sites. Learning about a system can also take place when managing practically. The adaptive management framework (Walters 1986) formally acknowledges this fact by evaluating learning in terms of how it will improve management of the system and therefore future system benefit. This is taken into account when ranking actions using stochastic dynamic programming (SDP). However, the benefits of any management action lie on a spectrum from pure system benefit, when there is nothing to be learned about the system, to pure knowledge gain. The current adaptive management framework does not permit management objectives to evaluate actions over the full range of this spectrum. By evaluating knowledge gain in units distinct to future system benefit this whole spectrum of management objectives can be unlocked. This paper outlines six decision making policies that differ across the spectrum of pure system benefit through to pure learning. The extensions to adaptive management presented allow specification of the relative importance of learning compared to system benefit in management objectives. Such an extension means practitioners can be more specific in the construction of conservation project objectives and be able to create policies for experimental management sites in the same framework as practical management sites.

Temporal

An evolving meditation upon the complex, periodic processes that mark Australia’s seasonality, and our increasing ability to disturb them. By amplifying and shining light upon a myriad of mysterious lives lived in blackness, the work presents a sensuous, deep engagement with the rich, irregular spectras of seasonal forms: whilst hinting at a far less comforting background increasingly framed by anthropogenic climate change. ’Temporal’ uses custom interactive systems, illusionary techniques and real time spatial audio processes that draw upon a rich array of media, including seasonal, nocturnal field recordings sourced in the Bundaberg region and detailed observations of foliage & flowering phases from that region. By drawing inspiration from the subtle transitions between what Europeans once named ‘Summer’ and ‘Autumn’ and the multiple seasons recognised by other cultures, whilst also including bodily disturbances within the work, ’Temporal’ creates a compellingly immersive environment that wraps audiences in luscious yet ominous atmospheres beyond sight and hearing. This work completes a two year long project of dynamic mediated installations that have been presented in Sydney, Beijing, Cairns and Bundanon, that have each been somehow choreographed by environmental cycles; alluding to a new framework for making works that we named ‘Seasonal’. These powerful, responsive & experiential works each draw attention to that which will disappear when biodiverse worlds have descended into an era of permanent darkness – an ‘extinction of human experience’. By tapping into the deeply interlocking seasonal cycles of environments that are themselves intimately linked with social, geographical & political concerns, participating audiences are therefore challenged to see the night, their locality & ecologies in new ways through extending their personal limits of perception, imagery & comprehension.

Proceedings of the Digital Cities 9 Workshop - Hackable Cities – From Subversive City Making to Systemic Change (Workshop) [Edited Proceedings]

The DC9 workshop takes place on June 27, 2015 in Limerick, Ireland and is titled “Hackable Cities: From Subversive City Making to Systemic Change”. The notion of “hacking” originates from the world of media technologies but is increasingly often being used for creative ideals and practices of city making. “City hacking” evokes more participatory, inclusive, decentralized, playful and subversive alternatives to often top-down ICT implementations in smart city making. However, these discourses about “hacking the city” are used ambiguously and are loaded with various ideological presumptions, which makes the term also problematic. For some “urban hacking” is about empowering citizens to organize around communal issues and perform aesthetic urban interventions. For others it raises questions about governance: what kind of “city hacks” should be encouraged or not, and who decides? Can city hacking be curated? For yet others, trendy participatory buzzwords like these are masquerades for deeply libertarian neoliberal values. Furthermore, a question is how “city hacking” may mature from the tactical level of smart and often playful interventions to the strategic level of enduring impact. The Digital Cities 9 workshop welcomes papers that explore the idea of “hackable city making” in constructive and critical ways.

Content-independent embedding scheme for multi-modal medical image watermarking

Background As the increasing adoption of information technology continues to offer better distant medical services, the distribution of, and remote access to digital medical images over public networks continues to grow significantly. Such use of medical images raises serious concerns for their continuous security protection, which digital watermarking has shown great potential to address. Methods We present a content-independent embedding scheme for medical image watermarking. We observe that the perceptual content of medical images varies widely with their modalities. Recent medical image watermarking schemes are image-content dependent and thus they may suffer from inconsistent embedding capacity and visual artefacts. To attain the image content-independent embedding property, we generalise RONI (region of non-interest, to the medical professionals) selection process and use it for embedding by utilising RONI’s least significant bit-planes. The proposed scheme thus avoids the need for RONI segmentation that incurs capacity and computational overheads. Results Our experimental results demonstrate that the proposed embedding scheme performs consistently over a dataset of 370 medical images including their 7 different modalities. Experimental results also verify how the state-of-the-art reversible schemes can have an inconsistent performance for different modalities of medical images. Our scheme has MSSIM (Mean Structural SIMilarity) larger than 0.999 with a deterministically adaptable embedding capacity. Conclusions Our proposed image-content independent embedding scheme is modality-wise consistent, and maintains a good image quality of RONI while keeping all other pixels in the image untouched. Thus, with an appropriate watermarking framework (i.e., with the considerations of watermark generation, embedding and detection functions), our proposed scheme can be viable for the multi-modality medical image applications and distant medical services such as teleradiology and eHealth.

Making things happen: How the environment triggers action

Innovative research conducted by Dr Jan Golembiewski from Medical Architectures Australasia-Pacific, in the area of environmental design for mental healthcare helps us reflect, through experiential storytelling, on the importance of paying attention to positive environmental ‘triggers for action’ when designing physical spaces to help manage particular patients’ conditions. This article is written about the mental health experience specifically, however environmental design is an issue we can all relate to in our everyday lives (at home, at work, social spots or in places such as hospitals or retreats) when we instinctively notice a connection between our mental health and our surroundings or a healthy sense of place.

A performance that requires two people

This work was a performance piece that took place at West Space as part of the 'Conceted Efforts' exhibition. For three hours, Antoinette J. Citizen and Courtney Coombs listed activities that require two people. The resulting list then remained in the gallery as an installed object. The work explores the role of collaboration in art practice as well as society more broadly.

The purpose and place of mentoring for women managers in organisations: An Australian perspective

Mentoring has been identified as an important career development activity for women managers. Over the last four decades, organisations world-wide have implemented formal mentoring programs for women and for members of minority groups in recognition of the personal and professional benefits mentoring provides, not only for persons who are mentored but also, for those who undertake the mentoring. This chapter reviews the literature on mentoring and the contribution it can make to the career development of women managers. It reviews several inter-related bodies of literature: women’s representation in management positions worldwide; theories and frameworks of mentoring; empirical research exploring the impact of mentoring relationships on women managers’ career development; current illustrations of formal programs offered to women managers in the public and private sectors; and some critical issues that continue to impact women managers in relation to mentoring relationships. The chapter concludes by making an argument for further research on gender and mentoring.

Genetic determinants of bone density and fracture risk: State of the art and future directions

Context: Osteoporosis is a common, highly heritable condition that causes substantial morbidity and mortality, the etiopathogenesis of which is poorly understood. Genetic studies are making increasingly rapid progress in identifying the genes involved. Evidence Acquisition and Synthesis: In this review, we will summarize the current understanding of the genetics of osteoporosis based on publications from PubMed from the year 1987 onward. Conclusions: Most genes involved in osteoporosis identified to date encode components of known pathways involved in bone synthesis or resorption, but as the field progresses, new pathways are being identified. Only a small proportion of the total genetic variation involved in osteoporosis has been identified, and new approaches will be required to identify most of the remaining genes.

The ninth 48hr game making challenge (2015)

The 48 hour game making challenge has been running since 2007. In recent years, we have not only been running a 'game jam' for the local community but we have also been exploring the way in which the event itself and the place of the event has the potential to create its own stories. The 2015 challenge is part of a series of data collection opportunities focussed on the game jam itself and the meaning making that the participants engage in about the event. We are continuing the data collection commenced in 2012: "Game jams are the creative festivals of the game development community and a game jam is very much an event or performance; its stories are those of subjective experience. Participants return year after year and recount personal stories from previous challenges; arrival in the 48hr location typically inspires instances of individual memory and narration more in keeping with those of a music festival or an oft frequented holiday destination. Since its inception, the 48hr has been heavily documented, from the photo-blogging of our first jam and the twitter streams of more recent events to more formal interviews and documentaries (see Anderson, 2012). We have even had our own moments of Gonzo journalism with an on-site press room one year and an ‘embedded’ journalist another year (Keogh, 2011). In the last two years of the 48hr we have started to explore ways and means to collect more abstract data during the event, that is, empirical data about movement and activity. The intent behind this form of data collection was to explore graphic and computer generated visualisations of the event, not for the purpose of formal analysis but in the service of further story telling." [exerpt from truna aka j.turner, Thomas & Owen, 2013) See: truna aka j.turner, Thomas & Owen (2013) Living the indie life: mapping creative teams in a 48 hour game jam and playing with data, Proceedings of the 9th Australasian Conference on Interactive Entertainment, IE'2013, September 30 - October 01 2013, Melbourne, VIC, Australia