OpenFABMAP : an open source toolbox for appearance-based loop closure detection

Appearance-based loop closure techniques, which leverage the high information content of visual images and can be used independently of pose, are now widely used in robotic applications. The current state-of-the-art in the field is Fast Appearance-Based Mapping (FAB-MAP) having been demonstrated in several seminal robotic mapping experiments. In this paper, we describe OpenFABMAP, a fully open source implementation of the original FAB-MAP algorithm. Beyond the benefits of full user access to the source code, OpenFABMAP provides a number of configurable options including rapid codebook training and interest point feature tuning. We demonstrate the performance of OpenFABMAP on a number of published datasets and demonstrate the advantages of quick algorithm customisation. We present results from OpenFABMAP’s application in a highly varied range of robotics research scenarios.

'The Pianist' and the ruins of Berlin

In this paper I will discuss some of the ways Berlin and the city's main studio complex, Studio Babelsberg, have been promoted and used as sites for international feature film production in recent years. I will use Roman Polanski's film The Pianist, which was shot in part at Studio Babelsberg and in the vicinity of Berlin, to exemplify some of the transformations and discuss some of the repercussions of international production for thinking about cinematic rivalry between places.

Melbourne on film dossier: 'Ghost Rider'

Despite being set in an unnamed Texan city, Ghost Rider (Mark Steven Johnson, 2007) was a landmark film for Melbourne. It was the first international production to be made at the Central City (now Docklands) Studios, much to the relief of the heavily-invested state government. And it demonstrated to the world (or more importantly, to producers resident in a small part of southern California), that the city was willing and available to be made over and made up to fit even (especially!) the most stupid filmmaking fantasies. Announcing the imminent arrival of the production in October 2004, the Herald Sun boldly predicted that '[c]ity lanes, Telstra Dome and the Yarra River will be the stars of the new action film'. In fact Melbourne is, as the filmmakers intended, (virtually) unidentifiable onscreen. For the Victorian State government, Ausfilm, Film Victoria and the Melbourne Film Office, this lack of specificity was something to be celebrated; a few months before Ghost Rider went in to production, the then State Minister for Innovation, John Brumby, declared that 'it's almost inconceivable that [Ghost Rider] won't put Melbourne on the map internationally'.

Cinematic hooks for Korean studies: using the 'Apache' framework for inspiring students about Korea in and through film

Developing awareness of and maintaining interest in Korea and Korean culture for non-language secondary and tertiary students continues to challenge educators in Australia. A lack of appropriate and accessible creative and cultural materials is a key factor contributing to this challenge. In light of changes made to 'fair use' guidelines for the Digital Millennium Copyright Act in the United States in July 2010, and in order to prepare for a time in the near future when Australian copyright regulations might follow suit, this article offers a framework for utilizing film and digital media contents in the classroom. Case studies of the short digital animation film 'Birthday Boy' (2004) and the feature film The Divine Weapon (2008) are presented in order to illustrate new educational approaches to popular Korean films---the cinematic component of the 'Korean Wave' ('Hanryu' or 'Hallyu' in Korean). It is hoped that this work-in-progress will enable teachers to inspire students with limited language skills to learn more about Korean popular culture, history, and tradition as well as media, politics, and genre studies in dynamic ways through the use of films as cultural texts in the classroom.

Secure authentication procedure with mobile device for ubiquitous health environments

Medical industries have brought Information Technology (IT) in their systems for both patients and medical staffs due to the numerous benefits of IT we experience at presently. Moreover, the Mobile healthcare (M-health) system has been developed as the first step of Ubiquitous Health Environment (UHE). With the mobility and multi-functions, M-health system will be able to provide more efficient and various services for both doctors and patients. Due to the invisible feature of mobile signals, hackers have easier access to hospital networks than wired network systems. This may result in several security incidents unless security protocols are well implemented. In this paper, user authentication and authorization procedures will applied as a featured component at each level of M-health systems inthe hospital environment. Accordingly, M-health system in the hospital will meet the optimal requirements as a countermeasure to its vulnerabilities.

Molecules, morphology, and ecology indicate a recent, amphibious ancestry for echidnas

The semiaquatic platypus and terrestrial echidnas (spiny anteaters) are the only living egg-laying mammals (monotremes). The fossil record has provided few clues as to their origins and the evolution of their ecological specializations; however, recent reassignment of the Early Cretaceous Teinolophos and Steropodon to the platypus lineage implies that platypuses and echidnas diverged >112.5 million years ago, reinforcing the notion of monotremes as living fossils. This placement is based primarily on characters related to a single feature, the enlarged mandibular canal, which supplies blood vessels and dense electrosensory receptors to the platypus bill. Our reevaluation of the morphological data instead groups platypus and echidnas to the exclusion of Teinolophos and Steropodon and suggests that an enlarged mandibular canal is ancestral for monotremes (partly reversed in echidnas, in association with general mandibular reduction). A multigene evaluation of the echidna–platypus divergence using both a relaxed molecular clock and direct fossil calibrations reveals a recent split of 19–48 million years ago. Platypus-like monotremes (Monotrematum) predate this divergence, indicating that echidnas had aquatically foraging ancestors that reinvaded terrestrial ecosystems. This ecological shift and the associated radiation of echidnas represent a recent expansion of niche space despite potential competition from marsupials. Monotremes might have survived the invasion of marsupials into Australasia by exploiting ecological niches in which marsupials are restricted by their reproductive mode. Morphology, ecology, and molecular biology together indicate that Teinolophos and Steropodon are basal monotremes rather than platypus relatives, and that living monotremes are a relatively recent radiation.

Private equity bidders : barbarians or best friends

The deal value of private equity merger and takeover activity has achieved unprecedented growth in the last couple of years, in Australia and globally. Private equity deals are not a new feature of the market; however, such deals have been subject to increased academic, professional and policy interest. This study examines the particular features of 15 major deals involving listed company "targets" and provides evidence – based on a comparison with a benchmark sample – to demonstrate the role that private equity plays in the market for corporate control. The objective of this study was to assess the friendliness of private equity bids. Based on the indicia compiled, lower bid premiums, the presence of break fees and the intention to retain senior management are compellingly different for private equity bids than for the comparative sample of bids. Using these several characteristics of "friendliness", the authors show that private equity deals are generally friendly in nature, consistent with industry rhetoric, but perhaps inconsistent with the popular belief that private equity bidders are the "barbarians at the gate".

Principals’ and teacher leaders’ perceptions of their role and interaction under current Chinese reform : a case study in Shandong province

This study aims to explore the perceptions of principals and teacher leaders regarding their roles and the interaction between these roles in Chinese urban primary schools at the time of unprecedented curriculum reform. This involves a complexity of factors such as the influence of globalisation, the impact of traditional Chinese cultural attitudes towards education through Confucianism, and the implementation requirements of the current education reforms. All of these wider contextual factors help to shape the leadership practices that are described in the study. A qualitative exploratory case study approach has been utilised to undertake this investigation. The conceptual framework for this study draws upon scholars‘ work from Western countries but has been adapted in order to address three research questions for the study‘s focus on the context in Shandong province, Mainland China. Three research questions were addressed: First, what are principals‘ perceptions of their leadership roles in Mainland China under current educational reform? Second, what are teacher leaders‘ perceptions of their leadership roles in Mainland China under current educational reform? And finally, what are principals‘ and teacher leaders‘ perceptions of how their roles interact? With reference to the principals in the study, the findings confirm Gurr‘s (2008) comprehensive leadership model relating to four roles, specifically, learning and teaching, symbolic and cultural awareness, future orientation, and accountability. Significantly, some sub-roles that emerge from the data are uniquely Chinese. For example, school culture construction is a very deliberate process in which principals and their staff talked openly about and were involved in creating a positive school climate comprising spiritual, material, and system dimensions. Another finding relates to school feature construction. This refers to the process that principals and staff used to make their schools distinctive and different from other schools and included such features as the school‘s philosophy and the school-based curriculum. In seeking to understand the nature of teacher leadership in Chinese primary schools, this research confirms some findings identified in Western literature. For instance, teacher leaders in Shandong province were involved in decision-making, working with parents and community members, undertaking and planning professional development for staff, and mediating between colleagues (Day & Harris, 2002; Harrison & Killion, 2007; Leithwood, Jantzi, & Steinbach, 1999; Muijs & Harris, 2006; Smylie, 1992). However, some new aspects, such as a heightened awareness of the importance of accountability, emerge from this study. The study‘s conceptual framework also draws upon some significant insights from micropolitics and, in particular, two core constructs, namely cooperation and conflict (Blase, 1991), to explore the interactions between principals and teacher leaders. In this study, principals and teacher leaders employed exchange and facilitation as two strategies in cooperative processes; and they adopted enforcement and compromise in conflictive processes. Finally, the study‘s findings indicate that principals and teacher leaders were developing new ways of interacting in response to the requirements of significant education reform. Most principals were exercising their power through (Blase, 1991) their teacher leaders who in turn, were working in alignment with their principals to achieve the desired outcomes in schools. It was significant that this form of 'parallel leadership' (Crowther, Ferguson, & Ham, 2009) characterised the teacher leadership roles at this period of change to the curriculum in Mainland China.

Gravy

To many aspiring writer/directors of feature film breaking into the industry may be perceived as an insurmountable obstacle. In contemplating my own attempt to venture into the world of feature filmmaking I have reasoned that a formulated strategy could be of benefit. As the film industry is largely concerned with economics I decided that writing a relatively low-cost feature film may improve my chances of being allowed directorship by a credible producer. As a result I have decided to write a modest feature film set in a single interior shooting location in an attempt to minimise production costs, therefore also attempting to reduce the perceived risk in hiring the writer as debut director. As a practice-led researcher, the primary focus of this research is to create a screenplay in response to my greater directorial aspirations and to explore the nature in which the said strategic decision to write a single-location film impacts on not only the craft of cinematic writing but also the creative process itself, as it pertains to the project at hand. The result is a comedy script titled Gravy, which is set in a single apartment and strives to maintain a fast comedic pace whilst employing a range of character and plot devices in conjunction with creative decisions that help to sustain cinematic interest within the confines of the apartment. In addition to the screenplay artifact, the exegesis also includes a section that reflects on the writing process in the form of personal accounts, decisions, problems and solutions as well as examination of other author’s works.

Hessian-based affine adaptation of salient local image features

Affine covariant local image features are a powerful tool for many applications, including matching and calibrating wide baseline images. Local feature extractors that use a saliency map to locate features require adaptation processes in order to extract affine covariant features. The most effective extractors make use of the second moment matrix (SMM) to iteratively estimate the affine shape of local image regions. This paper shows that the Hessian matrix can be used to estimate local affine shape in a similar fashion to the SMM. The Hessian matrix requires significantly less computation effort than the SMM, allowing more efficient affine adaptation. Experimental results indicate that using the Hessian matrix in conjunction with a feature extractor that selects features in regions with high second order gradients delivers equivalent quality correspondences in less than 17% of the processing time, compared to the same extractor using the SMM.

Unusual scene detection using distributed behaviour model and sparse representation

The ability to detect unusual events in surviellance footage as they happen is a highly desireable feature for a surveillance system. However, this problem remains challenging in crowded scenes due to occlusions and the clustering of people. In this paper, we propose using the Distributed Behavior Model (DBM), which has been widely used in computer graphics, for video event detection. Our approach does not rely on object tracking, and is robust to camera movements. We use sparse coding for classification, and test our approach on various datasets. Our proposed approach outperforms a state-of-the-art work which uses the social force model and Latent Dirichlet Allocation.

Discovering relevant features for effective query formulation

The quality of discovered features in relevance feedback (RF) is the key issue for effective search query. Most existing feedback methods do not carefully address the issue of selecting features for noise reduction. As a result, extracted noisy features can easily contribute to undesirable effectiveness. In this paper, we propose a novel feature extraction method for query formulation. This method first extract term association patterns in RF as knowledge for feature extraction. Negative RF is then used to improve the quality of the discovered knowledge. A novel information filtering (IF) model is developed to evaluate the proposed method. The experimental results conducted on Reuters Corpus Volume 1 and TREC topics confirm that the proposed model achieved encouraging performance compared to state-of-the-art IF models.

Visual route recognition with a handful of bits

In this paper we use a sequence-based visual localization algorithm to reveal surprising answers to the question, how much visual information is actually needed to conduct effective navigation? The algorithm actively searches for the best local image matches within a sliding window of short route segments or 'sub-routes', and matches sub-routes by searching for coherent sequences of local image matches. In contract to many existing techniques, the technique requires no pre-training or camera parameter calibration. We compare the algorithm's performance to the state-of-the-art FAB-MAP 2.0 algorithm on a 70 km benchmark dataset. Performance matches or exceeds the state of the art feature-based localization technique using images as small as 4 pixels, fields of view reduced by a factor of 250, and pixel bit depths reduced to 2 bits. We present further results demonstrating the system localizing in an office environment with near 100% precision using two 7 bit Lego light sensors, as well as using 16 and 32 pixel images from a motorbike race and a mountain rally car stage. By demonstrating how little image information is required to achieve localization along a route, we hope to stimulate future 'low fidelity' approaches to visual navigation that complement probabilistic feature-based techniques.

SeqSLAM : visual route-based navigation for sunny summer days and stormy winter nights

Learning and then recognizing a route, whether travelled during the day or at night, in clear or inclement weather, and in summer or winter is a challenging task for state of the art algorithms in computer vision and robotics. In this paper, we present a new approach to visual navigation under changing conditions dubbed SeqSLAM. Instead of calculating the single location most likely given a current image, our approach calculates the best candidate matching location within every local navigation sequence. Localization is then achieved by recognizing coherent sequences of these “local best matches”. This approach removes the need for global matching performance by the vision front-end - instead it must only pick the best match within any short sequence of images. The approach is applicable over environment changes that render traditional feature-based techniques ineffective. Using two car-mounted camera datasets we demonstrate the effectiveness of the algorithm and compare it to one of the most successful feature-based SLAM algorithms, FAB-MAP. The perceptual change in the datasets is extreme; repeated traverses through environments during the day and then in the middle of the night, at times separated by months or years and in opposite seasons, and in clear weather and extremely heavy rain. While the feature-based method fails, the sequence-based algorithm is able to match trajectory segments at 100% precision with recall rates of up to 60%.

Integrated vision-based attitude estimation

For many years, computer vision has lured researchers with promises of a low-cost, passive, lightweight and information-rich sensor suitable for navigation purposes. The prime difficulty in vision-based navigation is that the navigation solution will continually drift with time unless external information is available, whether it be cues from the appearance of the scene, a map of features (whether built online or known a priori), or from an externally-referenced sensor. It is not merely position that is of interest in the navigation problem. Attitude (i.e. the angular orientation of a body with respect to a reference frame) is integral to a visionbased navigation solution and is often of interest in its own right (e.g. flight control). This thesis examines vision-based attitude estimation in an aerospace environment, and two methods are proposed for constraining drift in the attitude solution; one through a novel integration of optical flow and the detection of the sky horizon, and the other through a loosely-coupled integration of Visual Odometry and GPS position measurements. In the first method, roll angle, pitch angle and the three aircraft body rates are recovered though a novel method of tracking the horizon over time and integrating the horizonderived attitude information with optical flow. An image processing front-end is used to select several candidate lines in a image that may or may not correspond to the true horizon, and the optical flow is calculated for each candidate line. Using an Extended Kalman Filter (EKF), the previously estimated aircraft state is propagated using a motion model and a candidate horizon line is associated using a statistical test based on the optical flow measurements and location of the horizon in the image. Once associated, the selected horizon line, along with the associated optical flow, is used as a measurement to the EKF. To evaluate the accuracy of the algorithm, two flights were conducted, one using a highly dynamic Uninhabited Airborne Vehicle (UAV) in clear flight conditions and the other in a human-piloted Cessna 172 in conditions where the horizon was partially obscured by terrain, haze and smoke. The UAV flight resulted in pitch and roll error standard deviations of 0.42° and 0.71° respectively when compared with a truth attitude source. The Cessna 172 flight resulted in pitch and roll error standard deviations of 1.79° and 1.75° respectively. In the second method for estimating attitude, a novel integrated GPS/Visual Odometry (GPS/VO) navigation filter is proposed, using a structure similar to a classic looselycoupled GPS/INS error-state navigation filter. Under such an arrangement, the error dynamics of the system are derived and a Kalman Filter is developed for estimating the errors in position and attitude. Through similar analysis to the GPS/INS problem, it is shown that the proposed filter is capable of recovering the complete attitude (i.e. pitch, roll and yaw) of the platform when subjected to acceleration not parallel to velocity for both the monocular and stereo variants of the filter. Furthermore, it is shown that under general straight line motion (e.g. constant velocity), only the component of attitude in the direction of motion is unobservable. Numerical simulations are performed to demonstrate the observability properties of the GPS/VO filter in both the monocular and stereo camera configurations. Furthermore, the proposed filter is tested on imagery collected using a Cessna 172 to demonstrate the observability properties on real-world data. The proposed GPS/VO filter does not require additional restrictions or assumptions such as platform-specific dynamics, map-matching, feature-tracking, visual loop-closing, gravity vector or additional sensors such as an IMU or magnetic compass. Since no platformspecific dynamics are required, the proposed filter is not limited to the aerospace domain and has the potential to be deployed in other platforms such as ground robots or mobile phones.