Locating people in video from semantic descriptions : a new database and approach

The location of previously unseen and unregistered individuals in complex camera networks from semantic descriptions is a time consuming and often inaccurate process carried out by human operators, or security staff on the ground. To promote the development and evaluation of automated semantic description based localisation systems, we present a new, publicly available, unconstrained 110 sequence database, collected from 6 stationary cameras. Each sequence contains detailed semantic information for a single search subject who appears in the clip (gender, age, height, build, hair and skin colour, clothing type, texture and colour), and between 21 and 290 frames for each clip are annotated with the target subject location (over 11,000 frames are annotated in total). A novel approach for localising a person given a semantic query is also proposed and demonstrated on this database. The proposed approach incorporates clothing colour and type (for clothing worn below the waist), as well as height and build to detect people. A method to assess the quality of candidate regions, as well as a symmetry driven approach to aid in modelling clothing on the lower half of the body, is proposed within this approach. An evaluation on the proposed dataset shows that a relative improvement in localisation accuracy of up to 21 is achieved over the baseline technique.

Empirical modelling of rolling shutter effect

We propose and evaluate a novel methodology to identify the rolling shutter parameters of a real camera. We also present a model for the geometric distortion introduced when a moving camera with a rolling shutter views a scene. Unlike previous work this model allows for arbitrary camera motion, including accelerations, is exact rather than a linearization and allows for arbitrary camera projection models, for example fisheye or panoramic. We show the significance of the errors introduced by a rolling shutter for typical robot vision problems such as structure from motion, visual odometry and pose estimation.

A hierarchical model of goal directed navigation selects trajectories in a visual environment

We have developed a Hierarchical Look-Ahead Trajectory Model (HiLAM) that incorporates the firing pattern of medial entorhinal grid cells in a planning circuit that includes interactions with hippocampus and prefrontal cortex. We show the model’s flexibility in representing large real world environments using odometry information obtained from challenging video sequences. We acquire the visual data from a camera mounted on a small tele-operated vehicle. The camera has a panoramic field of view with its focal point approximately 5 cm above the ground level, similar to what would be expected from a rat’s point of view. Using established algorithms for calculating perceptual speed from the apparent rate of visual change over time, we generate raw dead reckoning information which loses spatial fidelity over time due to error accumulation. We rectify the loss of fidelity by exploiting the loop-closure detection ability of a biologically inspired, robot navigation model termed RatSLAM. The rectified motion information serves as a velocity input to the HiLAM to encode the environment in the form of grid cell and place cell maps. Finally, we show goal directed path planning results of HiLAM in two different environments, an indoor square maze used in rodent experiments and an outdoor arena more than two orders of magnitude larger than the indoor maze. Together these results bridge for the first time the gap between higher fidelity bio-inspired navigation models (HiLAM) and more abstracted but highly functional bio-inspired robotic mapping systems (RatSLAM), and move from simulated environments into real-world studies in rodent-sized arenas and beyond.

Practical visual odometry for car-like vehicles

A method for calculating visual odometry for ground vehicles with car-like kinematic motion constraints similar to Ackerman's steering model is presented. By taking advantage of this non-holonomic driving constraint we show a simple and practical solution to the odometry calculation by clever placement of a single camera. The method has been implemented successfully on a large industrial forklift and a Toyota Prado SUV. Results from our industrial test site is presented demonstrating the applicability of this method as a replacement for wheel encoder-based odometry for these vehicles.

Just dawn

Commissioned for the It’s Timely exhibition at the Blacktown Arts Centre, Just Dawn is a response to two speeches that former Australian Prime Minister Gough Whitlam delivered in Blacktown in 1972 and 1974. Throughout the video, a series of white words and phrases fade in and out as a virtual camera flies towards an abstract horizon line. The narrative thread of the text is directed towards an unnamed Whitlam through the repeated appearance of the words ‘you said’. As the video progresses, the colours of the animated background slowly brighten to resemble an emerging dawn, and the sound, text and camera movements build in frequency and intensity. As they do so, the once optimistic outlook becomes increasingly unsteady. In these ways, Just Dawn is equal parts homage and lament for the ideological acuity and ambition of Whitlam’s agenda. It explores how Whitlam’s words can become markers for the complexities of both his own specific transformative policies, and the character of the socially progressive movement more broadly.

Protein ingestion increases myofibrillar protein synthesis after concurrent exercise

PURPOSE: We determined the effect of protein supplementation on anabolic signaling and rates of myofibrillar and mitochondrial protein synthesis after a single bout of concurrent training. METHODS: Using a randomized cross-over design, 8 healthy males were assigned to experimental trials consisting of resistance exercise (8 × 5 leg extension, 80% 1-RM) followed by cycling (30 min at ~70% VO2peak) with either post-exercise protein (PRO: 25 g whey protein) or placebo (PLA) ingestion. Muscle biopsies were obtained at rest, 1 and 4 h post-exercise. RESULTS: Akt and mTOR phosphorylation increased 1 h after exercise with PRO (175-400%, P<0.01) and was different from PLA (150-300%, P<0.001). MuRF1 and Atrogin-1 mRNA were elevated post-exercise but were higher with PLA compared to PRO at 1 h (50-315%, P<0.05), while PGC-1α mRNA increased 4 h post-exercise (620-730%, P<0.001) with no difference between treatments. Post-exercise rates of myofibrillar protein synthesis increased above rest in both trials (75-145%, P <0.05) but were higher with PRO (67%, P<0.05) while mitochondrial protein synthesis did not change from baseline. CONCLUSION: Our results show that a concurrent training session promotes anabolic adaptive responses and increases metabolic/oxidative mRNA expression in skeletal muscle. Protein ingestion after combined resistance and endurance exercise enhances myofibrillar protein synthesis and attenuates markers of muscle catabolism and thus is likely an important nutritional strategy to enhance adaptation responses with concurrent training.

Reduced resting skeletal muscle protein synthesis is rescued by resistance exercise and protein ingestion following short-term energy deficit

The myofibrillar protein synthesis (MPS) response to resistance exercise (REX) and protein ingestion during energy deficit (ED) is unknown. We determined, in young men (n=8) and women (n=7), protein signaling, resting post-absorptive MPS during energy balance [EB: 45 kcal∙(kg FFM∙d)-1] and after 5d of ED [30 kcal∙(kg FFM∙d)-1] as well as MPS while in ED after acute REX in the fasted state and with the ingestion of whey protein (15 and 30 g). Post-absorptive rates of MPS were 27% lower in ED than EB (P<0.001), but REX stimulated MPS to rates equal to EB. Ingestion of 15 and 30 g of protein after REX in ED increased MPS ~16 and ~34% above resting EB, (P<0.02). p70 S6Kthr389 phosphorylation increased above EB only with combined exercise and protein intake (~2-7 fold; P<0.05). In conclusion, short-term ED reduces post-absorptive MPS, however, a bout of REX in ED restores MPS to values observed at rest in EB. The ingestion of protein after REX further increases MPS above resting EB in a dose-dependent manner. We conclude that combining REX with increased protein availability after exercise enhances rates of skeletal muscle protein synthesis during short term ED and could, in the long term, preserve muscle mass.

Social turkey

Social Turkey is a short digital dance video developed over a series of weekly meetings with a group of sprightly and defiant sixty-plus year olds resident in Limerick. Very early in the process artist Ciara Finnegan and choreographer Jenny Roche found each individual expressing a wonderful enthusiasm for dance, an eagerness to perform and a healthy refusal to conform to stereotypes of aging. Finnegan was keen that this project should support an exchange of ideas rather than employ a top-down directorial structure. While Roche devised the fundamentals of the dance and Finnegan manned the camera, each participant contributed thereafter - improvising on a step sequence and collaborating on patterns that ultimately determined much of the look of the result. The work seeks to amplify the represented interests of a wider community while celebrating the vivacity of the particular group and the sheer fun of the collaboration.

Unusual event detection in crowded scenes

Novel computer vision techniques have been developed to automatically detect unusual events in crowded scenes from video feeds of surveillance cameras. The research is useful in the design of the next generation intelligent video surveillance systems. Two major contributions are the construction of a novel machine learning model for multiple instance learning through compressive sensing, and the design of novel feature descriptors in the compressed video domain.

A glossary of bodily responses

A collaborative video with Avril Huddy where two viewpoints, performer and documenter, were presented simultaneously to investigate Arakawa and Gin’s notion of “boundary-swaying”. In this performance-work, the performer influences what the camera is able to capture by engaging the documenter in a form of improvised dance. The performer’s movements appear impulsive and unpredictable, testing ways for the documenter to frame the performer’s movement. The images revealed by the documenter’s camera reflect a complexity of moments and co-incidences, evoking a sense of the performer’s embodied thinking within improvised movement. While a second camera uses a conventional wide angle shot to document the unfolding of the performance-work and track the connection between the documenter and the performer. While the performance-work itself is still highly-experimental, the ideas underpinning this exploration suggest how future investigations integrating more sensitive technology such as motion capture and tracking devises may be investigated. This performance-work formed part of the Creative Response Exhibition curated by Alan Prohm, Bill Lavender and Jason Nelson and a peer-review committee as part of the proceedings of the AG3 Online: Third International Arakawa and Gins Architecture and Philosophy Conference, hosted by the Centre for Public Culture and Ideas at Griffith University.

Stepping accuracy and visuomotor control among older adults : effect of target contrast and refractive blur

Purpose: Older adults have increased visual impairment, including refractive blur from presbyopic multifocal spectacle corrections, and are less able to extract visual information from the environment to plan and execute appropriate stepping actions; these factors may collectively contribute to their higher risk of falls. The aim of this study was to examine the effect of refractive blur and target visibility on the stepping accuracy and visuomotor stepping strategies of older adults during a precision stepping task. Methods: Ten healthy, visually normal older adults (mean age 69.4 ± 5.2 years) walked up and down a 20 m indoor corridor stepping onto selected high and low-contrast targets while viewing under three visual conditions: best-corrected vision, +2.00 DS and +3.00 DS blur; the order of blur conditions was randomised between participants. Stepping accuracy and gaze behaviours were recorded using an eyetracker and a secondary hand-held camera. Results: Older adults made significantly more stepping errors with increasing levels of blur, particularly exhibiting under-stepping (stepping more posteriorly) onto the targets (p<0.05), while visuomotor stepping strategies did not significantly alter. Stepping errors were also significantly greater for the low compared to the high contrast targets and differences in visuomotor stepping strategies were found, including increased duration of gaze and increased interval between gaze onset and initiation of the leg swing when stepping onto the low contrast targets. Conclusions: These findings highlight that stepping accuracy is reduced for low visibility targets, and for high levels of refractive blur at levels typically present in multifocal spectacle corrections, despite significant changes in some of the visuomotor stepping strategies. These findings highlight the importance of maximising the contrast of objects in the environment, and may help explain why older adults wearing multifocal spectacle corrections exhibit an increased risk of falling.

Design and test of a custom instrumented leg press for injury and recovery intervention

There is an increasing desire and emphasis to integrate assessment tools into the everyday training environment of athletes. These tools are intended to fine-tune athlete development, enhance performance and aid in the development of individualised programmes for athletes. The areas of workload monitoring, skill development and injury assessment are expected to benefit from such tools. This paper describes the development of an instrumented leg press and its application to testing leg dominance with a cohort of athletes. The developed instrumented leg press is a 45° reclining sled-type leg press with dual force plates, a displacement sensor and a CCD camera. A custom software client was developed using C#. The software client enabled near-real-time display of forces beneath each limb together with displacement of the quad track roller system and video feedback of the exercise. In recording mode, the collection of athlete particulars is prompted at the start of the exercise, and pre-set thresholds are used subsequently to separate the data into epochs from each exercise repetition. The leg press was evaluated in a controlled study of a cohort of physically active adults who performed a series of leg press exercises. The leg press exercises were undertaken at a set cadence with nominal applied loads of 50%, 100% and 150% of body weight without feedback. A significant asymmetry in loading of the limbs was observed in healthy adults during both the eccentric and concentric phases of the leg press exercise (P < .05). Mean forces were significantly higher beneath the non-dominant limb (4–10%) and during the concentric phase of the muscle action (5%). Given that symmetrical loading is often emphasized during strength training and remains a common goal in sports rehabilitation, these findings highlight the clinical potential for this instrumented leg press system to monitor symmetry in lower-limb loading during progressive strength training and sports rehabilitation protocols.

Biologically inspired SLAM using Wi-Fi

Wi-Fi is a commonly available source of localization information in urban environments but is challenging to integrate into conventional mapping architectures. Current state of the art probabilistic Wi-Fi SLAM algorithms are limited by spatial resolution and an inability to remove the accumulation of rotational error, inherent limitations of the Wi-Fi architecture. In this paper we leverage the low quality sensory requirements and coarse metric properties of RatSLAM to localize using Wi-Fi fingerprints. To further improve performance, we present a novel sensor fusion technique that integrates camera and Wi-Fi to improve localization specificity, and use compass sensor data to remove orientation drift. We evaluate the algorithms in diverse real world indoor and outdoor environments, including an office floor, university campus and a visually aliased circular building loop. The algorithms produce topologically correct maps that are superior to those produced using only a single sensor modality.

Fourier active appearance models

Gaining invariance to camera and illumination variations has been a well investigated topic in Active Appearance Model (AAM) fitting literature. The major problem lies in the inability of the appearance parameters of the AAM to generalize to unseen conditions. An attractive approach for gaining invariance is to fit an AAM to a multiple filter response (e.g. Gabor) representation of the input image. Naively applying this concept with a traditional AAM is computationally prohibitive, especially as the number of filter responses increase. In this paper, we present a computationally efficient AAM fitting algorithm based on the Lucas-Kanade (LK) algorithm posed in the Fourier domain that affords invariance to both expression and illumination. We refer to this as a Fourier AAM (FAAM), and show that this method gives substantial improvement in person specific AAM fitting performance over traditional AAM fitting methods.

Visual predictive control of spiral motion

This paper deals with constrained image-based visual servoing of circular and conical spiral motion about an unknown object approximating a single image point feature. Effective visual control of such trajectories has many applications for small unmanned aerial vehicles, including surveillance and inspection, forced landing (homing), and collision avoidance. A spherical camera model is used to derive a novel visual-predictive controller (VPC) using stability-based design methods for general nonlinear model-predictive control. In particular, a quasi-infinite horizon visual-predictive control scheme is derived. A terminal region, which is used as a constraint in the controller structure, can be used to guide appropriate reference image features for spiral tracking with respect to nominal stability and feasibility. Robustness properties are also discussed with respect to parameter uncertainty and additive noise. A comparison with competing visual-predictive control schemes is made, and some experimental results using a small quad rotor platform are given.