Activity modelling in crowded environments : a soft decision approach

Probabilistic topic models have recently been used for activity analysis in video processing, due to their strong capacity to model both local activities and interactions in crowded scenes. In those applications, a video sequence is divided into a collection of uniform non-overlaping video clips, and the high dimensional continuous inputs are quantized into a bag of discrete visual words. The hard division of video clips, and hard assignment of visual words leads to problems when an activity is split over multiple clips, or the most appropriate visual word for quantization is unclear. In this paper, we propose a novel algorithm, which makes use of a soft histogram technique to compensate for the loss of information in the quantization process; and a soft cut technique in the temporal domain to overcome problems caused by separating an activity into two video clips. In the detection process, we also apply a soft decision strategy to detect unusual events.We show that the proposed soft decision approach outperforms its hard decision counterpart in both local and global activity modelling.

An integrated approach for precise road reconstruction from aerial imagery and LiDAR data

Accurate and detailed road models play an important role in a number of geospatial applications, such as infrastructure planning, traffic monitoring, and driver assistance systems. In this thesis, an integrated approach for the automatic extraction of precise road features from high resolution aerial images and LiDAR point clouds is presented. A framework of road information modeling has been proposed, for rural and urban scenarios respectively, and an integrated system has been developed to deal with road feature extraction using image and LiDAR analysis. For road extraction in rural regions, a hierarchical image analysis is first performed to maximize the exploitation of road characteristics in different resolutions. The rough locations and directions of roads are provided by the road centerlines detected in low resolution images, both of which can be further employed to facilitate the road information generation in high resolution images. The histogram thresholding method is then chosen to classify road details in high resolution images, where color space transformation is used for data preparation. After the road surface detection, anisotropic Gaussian and Gabor filters are employed to enhance road pavement markings while constraining other ground objects, such as vegetation and houses. Afterwards, pavement markings are obtained from the filtered image using the Otsu's clustering method. The final road model is generated by superimposing the lane markings on the road surfaces, where the digital terrain model (DTM) produced by LiDAR data can also be combined to obtain the 3D road model. As the extraction of roads in urban areas is greatly affected by buildings, shadows, vehicles, and parking lots, we combine high resolution aerial images and dense LiDAR data to fully exploit the precise spectral and horizontal spatial resolution of aerial images and the accurate vertical information provided by airborne LiDAR. Objectoriented image analysis methods are employed to process the feature classiffcation and road detection in aerial images. In this process, we first utilize an adaptive mean shift (MS) segmentation algorithm to segment the original images into meaningful object-oriented clusters. Then the support vector machine (SVM) algorithm is further applied on the MS segmented image to extract road objects. Road surface detected in LiDAR intensity images is taken as a mask to remove the effects of shadows and trees. In addition, normalized DSM (nDSM) obtained from LiDAR is employed to filter out other above-ground objects, such as buildings and vehicles. The proposed road extraction approaches are tested using rural and urban datasets respectively. The rural road extraction method is performed using pan-sharpened aerial images of the Bruce Highway, Gympie, Queensland. The road extraction algorithm for urban regions is tested using the datasets of Bundaberg, which combine aerial imagery and LiDAR data. Quantitative evaluation of the extracted road information for both datasets has been carried out. The experiments and the evaluation results using Gympie datasets show that more than 96% of the road surfaces and over 90% of the lane markings are accurately reconstructed, and the false alarm rates for road surfaces and lane markings are below 3% and 2% respectively. For the urban test sites of Bundaberg, more than 93% of the road surface is correctly reconstructed, and the mis-detection rate is below 10%.

Unravel and picking at scabs : the underside of grief (a novella and exegesis)

This thesis consists of a 46,000 word polyphonic novella, Unravel, and an exegesis, Picking at Scabs: the Underside of Grief. The works are companion pieces, sitting side-by-side, and together they plumb the complex depths of loss and its resultant disorder, painful longing, and sorrow. The novella, representing 75% of the work and creative practice, is a multilayered work, which scrapes at the potent unspeakability of the presence of absence in the lives of its chief protagonists, Hana and Guy. As the novella progresses, loss is unraveled to reveal the interplay of remembering and forgetting, past and present and the ways in which these knotty fibres are connected with the strands of memory, trauma, silence, and the uncanny. Each of these threads is woven into the novella and as they plait together, loosen and fray, they expose the mystery, lies and secrets at the core of the novella. The exegesis, which comprises 25% of the thesis, picks at loss to uncover and loosen a complex and worn tangle of knots and loops. In this way, the exegesis and creative work are constantly in dialogue and while neither provides all the answers, both stretch the yarn to reveal an enthusiasm of practice.

Some reflections about P2M : the place of the mirror

The purpose of this paper is to provide some insights about P2M, and more specifically, to develop some thoughts about Project Management seen as a Mirror, a place for reflection…, between the Mission of organisation and its actual creation of Values (with s: a source of value for people, organisations and society). This place is the realm of complexity, of interactions between multiple variables, each of them having a specific time horizon and occupying a specific place, playing a specific role. Before developing this paper I would like to borrow to my colleague and friend, Professor Ohara, the following, part of a paper going to be presented at IPMA World Congress, in New Delhi later this year in November 2005. “P2M is the Japanese version of project & program management, which is the first standard guide for education and certification developed in 2001. A specific finding of P2M is characterized by “mission driven management of projects” or a program which harness complexity of problem solving observed in the interface between technical system and business model.” (Ohara, 2005, IPMA Conference, New Delhi) “The term of “mission” is a key word in the field of corporate strategy, where it expresses raison d’être or “value of business”. It is more specifically used for expressing “the client needs” in terms of a strategic business unit. The concept of mission is deemed to be a useful tool to share essential content of value and needs in message for complex project.” (Ohara, 2005, IPMA Conference, New Delhi) “Mission is considered as a significant “metamodel representation” by several reasons. First, it represents multiple values for aspiration. The central objective of mission initiative is profiling of ideality in the future from reality, which all stakeholders are glad to accept and share. Second, it shall be within a stretch of efforts, and not beyond or outside of the realization. Though it looks like unique, it has to depict a solid foundation. The pragmatic sense of equilibrium between innovation and adaptation is required for the mission. Third, it shall imply a rough sketch for solution to critical issues for problems in reality.” (Ohara, 2005, IPMA Conference, New Delhi) “Project modeling” idea has been introduced in P2M program management. A package of three project models of “scheme”, “system” and “service” are given as a reference type program. (Ohara, 2005, IPMA Conference, New Delhi) If these quotes apply to P2M, they are fully congruent with the results of the research undertaken and the resulting meta-model & meta-method developed by the CIMAP, ESC Lille Research Centre in Project & Program Management, since the 80’s. The paper starts by questioning the common Project Management (PM) paradigm. Then discussing the concept of Project, it argues that an alternative epistemological position should be taken to capture Page 2 / 11 the very nature of the PM field. Based on this, a development about “the need of modelling to understand” is proposed grounded on two theoretical roots. This leads to the conclusion that, in order to enables this modelling, a standard approach is necessary, but should be understood under the perspective of the Theory of Convention in order to facilitate a situational and contextual application.

Something rotten in the state of Minnesota: or the morality of backwoodsmen: 'A Simple Plan'

'A Simple Plan' is a deceptively complex and multilayered film, combining elements of Celtic mythology with the morality play and the windfall fantasy gone disastrously wrong. Despite its blending of realism and heavyhanded symbolism, and its abundant trans-textual gestures, 'A Simple Plan' is in many ways defiantly not a 90s movie: its leading characters are fashionably flawed, but they are neither sensitive, nor honourable, nor heroic; there are no startling special effects or intricate timeshifts; and it desperately gives the impression of depth, of being emphatically more than mere superficial excess. At a stretch it almost appears to be a throwback to the 1930s Production Code emphasis on the role of cinema in moral instruction; while good hardly triumphs over evil, venality is painfully and emphatically punished. But in other ways it is a quintessential late 90s film: an American/British/Japanese/German/French co-production, 'A Simple Plan' acts most palpably as a commentary on the moral, economic and social condition of the United States at the end of the American century.

Outward-looking Australian cinema

Over the last twenty years or so, Australian cinema's international relations in production and policy have expanded and become more complex, while those with Hollywood have been transformed. The boundaries of the national cinema stretch much further than the national territory. Australian production and postproduction companies work in Australia with international partners or on international projects. In this article I will trace some of the material and discursive entailments of this new international turn to explore how dynamic and shifting relations between the local/national and the international have transformed the ways in which we might think about what constitutes Australian cinema, and to illustrate how relations of commonality and continuity with the international called up in the new arrangements challenge the dominant articulation in policy of difference from 'other kinds of filmmaking' as the basis of Australian cinema. I draw on Deb Verhoeven's work on simultaneously national and international films and filmmakers, and adapt Doreen Massey's concept of 'outwardlookingness' to consider Australian cinema's international aspects.

In the pursuit of effective affective computing : the relationship between features and registration

For facial expression recognition systems to be applicable in the real world, they need to be able to detect and track a previously unseen person's face and its facial movements accurately in realistic environments. A highly plausible solution involves performing a "dense" form of alignment, where 60-70 fiducial facial points are tracked with high accuracy. The problem is that, in practice, this type of dense alignment had so far been impossible to achieve in a generic sense, mainly due to poor reliability and robustness. Instead, many expression detection methods have opted for a "coarse" form of face alignment, followed by an application of a biologically inspired appearance descriptor such as the histogram of oriented gradients or Gabor magnitudes. Encouragingly, recent advances to a number of dense alignment algorithms have demonstrated both high reliability and accuracy for unseen subjects [e.g., constrained local models (CLMs)]. This begs the question: Aside from countering against illumination variation, what do these appearance descriptors do that standard pixel representations do not? In this paper, we show that, when close to perfect alignment is obtained, there is no real benefit in employing these different appearance-based representations (under consistent illumination conditions). In fact, when misalignment does occur, we show that these appearance descriptors do work well by encoding robustness to alignment error. For this work, we compared two popular methods for dense alignment-subject-dependent active appearance models versus subject-independent CLMs-on the task of action-unit detection. These comparisons were conducted through a battery of experiments across various publicly available data sets (i.e., CK+, Pain, M3, and GEMEP-FERA). We also report our performance in the recent 2011 Facial Expression Recognition and Analysis Challenge for the subject-independent task.

Quality based frame selection for video face recognition

Quality based frame selection is a crucial task in video face recognition, to both improve the recognition rate and to reduce the computational cost. In this paper we present a framework that uses a variety of cues (face symmetry, sharpness, contrast, closeness of mouth, brightness and openness of the eye) to select the highest quality facial images available in a video sequence for recognition. Normalized feature scores are fused using a neural network and frames with high quality scores are used in a Local Gabor Binary Pattern Histogram Sequence based face recognition system. Experiments on the Honda/UCSD database shows that the proposed method selects the best quality face images in the video sequence, resulting in improved recognition performance.

A three-nucleotide mutation altering the Maize streak virus Rep pRBR-interaction motif reduces symptom severity in maize and partially reverts at high frequency without restoring pRBR-Rep binding

Geminivirus infectivity is thought to depend on interactions between the virus replication-associated proteins Rep or RepA and host retinoblastoma-related proteins (pRBR), which control cell-cycle progression. It was determined that the substitution of two amino acids in the Maize streak virus (MSV) RepA pRBR-interaction motif (LLCNE to LLCLK) abolished detectable RepA-pRBR interaction in yeast without abolishing infectivity in maize. Although the mutant virus was infectious in maize, it induced less severe symptoms than the wild-type virus. Sequence analysis of progeny viral DNA isolated from infected maize enabled detection of a high-frequency single-nucleotide reversion of C(601)A in the 3 nt mutated sequence of the Rep gene. Although it did not restore RepA-pRBR interaction in yeast, sequence-specific PCR showed that, in five out of eight plants, the C(601)A reversion appeared by day 10 post-inoculation. In all plants, the C(601)A revertant eventually completely replaced the original mutant population, indicating a high selection pressure for the single-nucleotide reversion. Apart from potentially revealing an alternative or possibly additional function for the stretch of DNA that encodes the apparently non-essential pRBR-interaction motif of MSV Rep, the consistent emergence and eventual dominance of the C(601)A revertant population might provide a useful tool for investigating aspects of MSV biology, such as replication, mutation and evolution rates, and complex population phenomena, such as competition between quasispecies and population turnover. © 2005 SGM.

Numerical simulation of red blood cells' deformation using SPH method

A numerical simulation method for the Red Blood Cells’ (RBC) deformation is presented in this study. The two-dimensional RBC membrane is modeled by the spring network, where the elastic stretch/compression energy and the bending energy are considered with the constraint of constant RBC surface area. Smoothed Particle Hydrodynamics (SPH) method is used to solve the Navier-Stokes equation coupled with the Plasma-RBC membrane and Cytoplasm- RBC membrane interaction. To verify the method, the motion of a single RBC is simulated in Poiseuille flow and compared with the results reported earlier. Typical motion and deformation mechanism of the RBC is observed.

Socio-economic & environmental impacts of battery driven auto rickshaw at Rajshahi city in Bangladesh

This paper describes the socio-economic and environmental impacts of battery driven Auto Rickshaw at Rajshahi city in Bangladesh. Unemployment problem is one of the major problems in Bangladesh. The number of unemployed people in Bangladesh is 7 lacks. Auto Rickshaw reduces this unemployment problem near about 2%.In this thesis work various questions were asked to the Auto Rickshaw driver in the different point in the Rajshahi city. Then those data were calculated to know their socio economic condition. The average number of passenger per Auto Rickshaw was determined at various places of Rajshahi city (Talaimari mor, Hadir mor, Alupotti, Shaheb bazar zero point, Shodor Hospital mor, Fire brigade mor, CNB mor, Lakshipur mor, Bondo gate, Bornali, Panir tank, Rail gate, Rail Station, Bhodrar mor, Adorsha School mor). Air pollution is a great threat for human health. One of the major causes of the air pollution is the emission from various vehicles, which are running by the burning of the fossil fuel in different internal combustion(IC) engines. All the data’s about emission from various power plants were collected from internet. Then the amounts of emission (CO2, NOX and PM) from different power plant were calculated in terms of kg/km. The energy required by the Auto Rickshaw per km was also calculated. Then the histogram of emission from different vehicles in terms of kg/km was drawn. By analyzing the data and chart, it was found that, battery driven Auto Rickshaw increases income, social status, comfort and decreases unemployment problems.

Social signal processing for pain monitoring using a hidden conditional random field

utomatic pain monitoring has the potential to greatly improve patient diagnosis and outcomes by providing a continuous objective measure. One of the most promising methods is to do this via automatically detecting facial expressions. However, current approaches have failed due to their inability to: 1) integrate the rigid and non-rigid head motion into a single feature representation, and 2) incorporate the salient temporal patterns into the classification stage. In this paper, we tackle the first problem by developing a “histogram of facial action units” representation using Active Appearance Model (AAM) face features, and then utilize a Hidden Conditional Random Field (HCRF) to overcome the second issue. We show that both of these methods improve the performance on the task of pain detection in sequence level compared to current state-of-the-art-methods on the UNBC-McMaster Shoulder Pain Archive.

Second Skin 2012 Workshop

The Second Skin 2012 Workshop Program consisted of a full-day intensive design immersion workshop run on Saturday 14 July 2012, at the QUT Faculty of Creative Industries Fashion Studios at Kelvin Grove Brisbane, Australia, for 30? self-selected high-achieving junior and middle school (year 5-9) students, as part of the Queensland Academies ‘Young Scholars’ Program. Inspired by a scientist researching the impact of sun on skin, and mentored by tertiary fashion design and interior design educators, and six tertiary fashion design and interior design students, the workshop explored science and design-inspired prototype solutions for sun-safety. This action research study aimed to facilitate an acute awareness in young people of the sun safety message (alternative to a scare campaign), the role of design in society and the value of design thinking skills in solving complex challenges, and to inspire the generation of strategies to address a systemic health issue. It also aimed to investigate the value of collaboration between junior and middle school students, tertiary design educators and students and industry professionals in targeting youth sun safety, and inspiring post-secondary pathways and idea generation for education. During the workshop, students developed sketching, making, communication, presentation and collaboration skills to improve their design process, while considering social, cultural and environmental opportunities. Through a series of hands-on collaborative design experiments, participants explored in teams of five, ways in which a ‘second skin’ can mirror elements of our skin – the ability to protect, divide, enclose, stretch, scar, pattern, peel and reveal – inspiring both functional and aesthetic design solutions. Underpinned by the State Library of Queensland Design Minds Website ‘inquire, ideate and implement’ model of design thinking, the experiments culminated in the development of a detailed client brief, the design and fabrication of a fashionable sun safe clothing range and then a team presentation and modelling of prototypes in a fashion parade, viewed also by parents. The final collections were judged by three prominent judges: Louise Baldwin - Executive Manager Public Health QLD Cancer Council, Shane Thompson - Architect and 2012 Queensland Smart Design Fellow, and Leigh Buchanan – Fashion designer and Project Runway Australia finalist. The workshop was filmed for Queensland television program ‘Totally Wild’ for dissemination of the value of design, the Design Minds model and the sun safety message to a wider target youth audience.

The time course of in vivo recovery of transverse strain in high-stress tendons following exercise

Objective To evaluate the time course of the recovery of transverse strain in the Achilles and patellar tendon following a bout of resistance exercise. Methods Seventeen healthy adults underwent sonographic examination of the right patellar (n=9) and Achilles (n=8) tendons immediately prior to and following 90 repetitions of weight-bearing quadriceps and gastrocnemius-resistance exercise performed against an effective resistance of 175% and 250% body weight, respectively. Sagittal tendon thickness was determined 20 mm from the enthesis and transverse strain, as defined by the stretch ratio, was repeatedly monitored over a 24 h recovery period. Results Resistance exercise resulted in an immediate decrease in Achilles (t7=10.6, p<0.01) and patellar (t8=8.9, p<0.01) tendon thickness, resulting in an average transverse stretch ratio of 0.86±0.04 and 0.82±0.05, which was not significantly different between tendons. The magnitude of the immediate transverse strain response, however, was reduced with advancing age (r=0.63, p<0.01). Recovery in transverse strain was prolonged compared with the duration of loading and exponential in nature. The average primary recovery time was not significantly different between the Achilles (6.5±3.2 h) and patellar (7.1±3.2 h) tendons. Body weight accounted for 62% and 64% of the variation in recovery time, respectively. Conclusions Despite structural and biochemical differences between the Achilles and patellar tendon, the mechanisms underlying transverse creep recovery in vivo appear similar and are highly time dependent. These novel findings have important implications concerning the time required for the mechanical recovery of high-stress tendons following an acute bout of exercise.

Automated marine mammal detection from aerial imagery

This paper presents two algorithms to automate the detection of marine species in aerial imagery. An algorithm from an initial pilot study is presented in which morphology operations and colour analysis formed the basis of its working principle. A second approach is presented in which saturation channel and histogram-based shape profiling were used. We report on performance for both algorithms using datasets collected from an unmanned aerial system at an altitude of 1000 ft. Early results have demonstrated recall values of 48.57% and 51.4%, and precision values of 4.01% and 4.97%.