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.

Conclusion

This monograph provides an overview of recruitment learning approaches from a computational perspective. Recruitment learning is a unique machine learning technique that: (1) explains the physical or functional acquisition of new neurons in sparsely connected networks as a biologically plausible neural network method; (2) facilitates the acquisition of new knowledge to build and extend knowledge bases and ontologies as an artificial intelligence technique; (3) allows learning by use of background knowledge and a limited number of observations, consistent with psychological theory.

A randomised controlled trial of an online theory-based intervention to improve adult Australians' sun-protective behaviours

Objective. To evaluate the effectiveness of a single-session online theory of planned behaviour (TPB)-based intervention to improve sun-protective attitudes and behaviour among Australian adults. Methods. Australian adults (N = 534; 38.7% males; Mage = 39.3 years) from major cities (80.9%), regional (17.6%) and remote areas (1.5%)were recruited and randomly allocated to an intervention (N=265) and information only group (N = 267). The online intervention focused on fostering positive attitudes, perceptions of normative support, and control perceptions for sun protection. Participants completed questionnaires assessing standard TPB measures (attitude, subjective norm, perceived behavioural control, intention, behaviour) and extended TPB constructs of group norm (friends, family), personal norm, and image norm, pre-intervention (Time 1) and one week (Time 2) and one month post-intervention (Time 3). Repeated Measures Multivariate Analysis of Variance tested intervention effects across time. Results. Intervention participants reported more positive attitudes towards sun protection and used sunprotective measures more often in the subsequent month than participants receiving information only. The intervention effects on control perceptions and norms were non-significant. Conclusions. A theory-based online intervention fostering more favourable attitudes towards sun safety can increase sun protection attitudes and self-reported behaviour among Australian adults in the short term.

Representing team behaviours from noisy data using player role

Due to their unobtrusive nature, vision-based approaches to tracking sports players have been preferred over wearable sensors as they do not require the players to be instrumented for each match. Unfortunately however, due to the heavy occlusion between players, variation in resolution and pose, in addition to fluctuating illumination conditions, tracking players continuously is still an unsolved vision problem. For tasks like clustering and retrieval, having noisy data (i.e. missing and false player detections) is problematic as it generates discontinuities in the input data stream. One method of circumventing this issue is to use an occupancy map, where the field is discretised into a series of zones and a count of player detections in each zone is obtained. A series of frames can then be concatenated to represent a set-play or example of team behaviour. A problem with this approach though is that the compressibility is low (i.e. the variability in the feature space is incredibly high). In this paper, we propose the use of a bilinear spatiotemporal basis model using a role representation to clean-up the noisy detections which operates in a low-dimensional space. To evaluate our approach, we used a fully instrumented field-hockey pitch with 8 fixed high-definition (HD) cameras and evaluated our approach on approximately 200,000 frames of data from a state-of-the-art real-time player detector and compare it to manually labeled data.

Adaptive online gradient descent

We study the rates of growth of the regret in online convex optimization. First, we show that a simple extension of the algorithm of Hazan et al eliminates the need for a priori knowledge of the lower bound on the second derivatives of the observed functions. We then provide an algorithm, Adaptive Online Gradient Descent, which interpolates between the results of Zinkevich for linear functions and of Hazan et al for strongly convex functions, achieving intermediate rates between [square root T] and [log T]. Furthermore, we show strong optimality of the algorithm. Finally, we provide an extension of our results to general norms.

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

Do-Rank: DCG optimization for learning-to-rank in tag-based item recommendation systems

Discounted Cumulative Gain (DCG) is a well-known ranking evaluation measure for models built with multiple relevance graded data. By handling tagging data used in recommendation systems as an ordinal relevance set of {negative,null,positive}, we propose to build a DCG based recommendation model. We present an efficient and novel learning-to-rank method by optimizing DCG for a recommendation model using the tagging data interpretation scheme. Evaluating the proposed method on real-world datasets, we demonstrate that the method is scalable and outperforms the benchmarking methods by generating a quality top-N item recommendation list.

Is the sun too bright in Queensland? An approach to robust outdoor colour beacon detection

Using cameras onboard a robot for detecting a coloured stationary target outdoors is a difficult task. Apart from the complexity of separating the target from the background scenery over different ranges, there are also the inconsistencies with direct and reflected illumination from the sun,clouds, moving and stationary objects. They can vary both the illumination on the target and its colour as perceived by the camera. In this paper, we analyse the effect of environment conditions, range to target, camera settings and image processing on the reported colours of various targets. The analysis indicates the colour space and camera configuration that provide the most consistent colour values over varying environment conditions and ranges. This information is used to develop a detection system that provides range and bearing to detected targets. The system is evaluated over various lighting conditions from bright sunlight, shadows and overcast days and demonstrates robust performance. The accuracy of the system is compared against a laser beacon detector with preliminary results indicating it to be a valuable asset for long-range coloured target detection.

LunaRoo: A Proposal for the Google Lunar XPrize Payload Opportunity with the Part Time Scientists team

This proposal describes the innovative and competitive lunar payload solution developed at the Queensland University of Technology (QUT)–the LunaRoo: a hopping robot designed to exploit the Moon's lower gravity to leap up to 20m above the surface. It is compact enough to fit within a 10cm cube, whilst providing unique observation and mission capabilities by creating imagery during the hop. This first section is deliberately kept short and concise for web submission; additional information can be found in the second chapter.