Unsupervised speaker adaptation for telephone call transcription

The use of the PC and Internet for placing telephone calls will present new opportunities to capture vast amounts of un-transcribed speech for a particular speaker. This paper investigates how to best exploit this data for speaker-dependent speech recognition. Supervised and unsupervised experiments in acoustic model and language model adaptation are presented. Using one hour of automatically transcribed speech per speaker with a word error rate of 36.0%, unsupervised adaptation resulted in an absolute gain of 6.3%, equivalent to 70% of the gain from the supervised case, with additional adaptation data likely to yield further improvements. LM adaptation experiments suggested that although there seems to be a small degree of speaker idiolect, adaptation to the speaker alone, without considering the topic of the conversation, is in itself unlikely to improve transcription accuracy.

Treadmill adaptation and verification of self-selected walking speed : a protocol for children

The term self-selected (i.e., individual or comfortable walking pace or speed) is commonly used in the literature (Frost, Dowling, Bar-Or, & Dyson, 1997; Jeng, Liao, Lai, & Hou, 1997; Wergel-Kolmert & Wohlfart, 1999; Maltais, Bar-Or, Pienynowski, & Galea, 2003; Browning & Kram, 2005; Browning, Baker, Herron, & Kram, 2006; Hills, Byrne, Wearing, & Armstrong, 2006) and is identified as the most efficient walking speed, with increased efficiency defined by lower oxygen uptake (VO^sub 2^) per unit mechanical work (Hoyt & Taylor, 1981; Taylor, Heglund, & Maloiy, 1982; Hreljac, 1993). [...] assessing individual and group differences in metabolic energy expenditure using oxygen uptake requires individuals to be comfortable with, and able to accommodate to, the equipment.

Seeing pedestrians at night : visual clutter does not mask biological motion

Although placing reflective markers on pedestrians’ major joints can make pedestrians more conspicuous to drivers at night, it has been suggested that this “biological motion” effect may be reduced when visual clutter is present. We tested whether extraneous points of light affected the ability of 12 younger and 12 older drivers to see pedestrians as they drove on a closed road at night. Pedestrians wore black clothing alone or with retroreflective markings in four different configurations. One pedestrian walked in place and was surrounded by clutter on half of the trials. Another was always surrounded by visual clutter but either walked in place or stood still. Clothing configuration, pedestrian motion, and driver age influenced conspicuity but clutter did not. The results confirm that even in the presence of visual clutter pedestrians wearing biological motion configurations are recognized more often and at greater distances than when they wear a reflective vest.

A brush with the real world : the future of inertial motion capture in live performance

3D Motion capture is a medium that plots motion, typically human motion, converting it into a form that can be represented digitally. It is a fast evolving field and recent inertial technology may provide new artistic possibilities for its use in live performance. Although not often used in this context, motion capture has a combination of attributes that can provide unique forms of collaboration with performance arts. The inertial motion capture suit used for this study has orientation sensors placed at strategic points on the body to map body motion. Its portability, real-time performance, ease of use, and its immunity from line-of-sight problems inherent in optical systems suggest it would work well as a live performance technology. Many animation techniques can be used in real-time. This research examines a broad cross-section of these techniques using four practice-led cases to assess the suitability of inertial motion capture to live performance. Although each case explores different visual possibilities, all make use of the performativity of the medium, using either an improvisational format or interactivity among stage, audience and screen that would be difficult to emulate any other way. A real-time environment is not capable of reproducing the depth and sophistication of animation people have come to expect through media. These environments take many hours to render. In time the combination of what can be produced in real-time and the tools available in a 3D environment will no doubt create their own tree of aesthetic directions in live performance. The case study looks at the potential of interactivity that this technology offers.

Adaptation to global warming by changing internal loads of buildings

Global warming can have a significant impact on the building thermal environment and energy performance. Because greenhouse gas concentrations are still continuing to increase, this warming process will continue and may accelerate. Adaptation to global warming is therefore emerging as one of the key requirements for buildings. This requires all the existing and new buildings not only to perform and operate satisfactorily in the new environment but also to satisfy the environmental performance criteria of sustainability. Through a parametric study using the building simulation technique, this paper investigates the adaptation potential of changing the building internal load densities to the future global warming. Case studies for office buildings in major Australian capital cities are presented. Based on the results of parametric study, possible adaptation strategies are also proposed and evaluated.

Abandoned object detection using multi-layer motion detection

Abandoned object detection (AOD) systems are required to run in high traffic situations, with high levels of occlusion. Systems rely on background segmentation techniques to locate abandoned objects, by detecting areas of motion that have stopped. This is often achieved by using a medium term motion detection routine to detect long term changes in the background. When AOD systems are integrated into person tracking system, this often results in two separate motion detectors being used to handle the different requirements. We propose a motion detection system that is capable of detecting medium term motion as well as regular motion. Multiple layers of medium term (static) motion can be detected and segmented. We demonstrate the performance of this motion detection system and as part of an abandoned object detection system.

Object tracking using multiple motion modalities

This paper presents an object tracking system that utilises a hybrid multi-layer motion segmentation and optical flow algorithm. While many tracking systems seek to combine multiple modalities such as motion and depth or multiple inputs within a fusion system to improve tracking robustness, current systems have avoided the combination of motion and optical flow. This combination allows the use of multiple modes within the object detection stage. Consequently, different categories of objects, within motion or stationary, can be effectively detected utilising either optical flow, static foreground or active foreground information. The proposed system is evaluated using the ETISEO database and evaluation metrics and compared to a baseline system utilising a single mode foreground segmentation technique. Results demonstrate a significant improvement in tracking results can be made through the incorporation of the additional motion information.

Improved simultaneous computation of motion detection and optical flow for object tracking

Object tracking systems require accurate segmentation of the objects from the background for effective tracking. Motion segmentation or optical flow can be used to segment incoming images. Whilst optical flow allows multiple moving targets to be separated based on their individual velocities, optical flow techniques are prone to errors caused by changing lighting and occlusions, both common in a surveillance environment. Motion segmentation techniques are more robust to fluctuating lighting and occlusions, but don't provide information on the direction of the motion. In this paper we propose a combined motion segmentation/optical flow algorithm for use in object tracking. The proposed algorithm uses the motion segmentation results to inform the optical flow calculations and ensure that optical flow is only calculated in regions of motion, and improve the performance of the optical flow around the edge of moving objects. Optical flow is calculated at pixel resolution and tracking of flow vectors is employed to improve performance and detect discontinuities, which can indicate the location of overlaps between objects. The algorithm is evaluated by attempting to extract a moving target within the flow images, given expected horizontal and vertical movement (i.e. the algorithms intended use for object tracking). Results show that the proposed algorithm outperforms other widely used optical flow techniques for this surveillance application.

Inertial motion capture and live performance (with a focus on dance)

3D Motion capture is a fast evolving field and recent inertial technology may expand the artistic possibilities for its use in live performance. Inertial motion capture has three attributes that make it suitable for use with live performance; it is portable, easy to use and can operate in real-time. Using four projects, this paper discusses the suitability of inertial motion capture to live performance with a particular emphasis on dance. Dance is an artistic application of human movement and motion capture is the means to record human movement as digital data. As such, dance is clearly a field in which the use of real-time motion capture is likely to become more common, particularly as projected visual effects including real-time video are already often used in dance performances. Understandably, animation generated in real-time using motion capture is not as extensive or as clean as the highly mediated animation used in movies and games, but the quality is still impressive and the ‘liveness’ of the animation has compensating features that offer new ways of communicating with an audience.

Unmanned Aerial Vehicles UAVs attitude, height, motion estimation and control using visual systems

This paper presents an implementation of an aircraft pose and motion estimator using visual systems as the principal sensor for controlling an Unmanned Aerial Vehicle (UAV) or as a redundant system for an Inertial Measure Unit (IMU) and gyros sensors. First, we explore the applications of the unified theory for central catadioptric cameras for attitude and heading estimation, explaining how the skyline is projected on the catadioptric image and how it is segmented and used to calculate the UAV’s attitude. Then we use appearance images to obtain a visual compass, and we calculate the relative rotation and heading of the aerial vehicle. Additionally, we show the use of a stereo system to calculate the aircraft height and to measure the UAV’s motion. Finally, we present a visual tracking system based on Fuzzy controllers working in both a UAV and a camera pan and tilt platform. Every part is tested using the UAV COLIBRI platform to validate the different approaches, which include comparison of the estimated data with the inertial values measured onboard the helicopter platform and the validation of the tracking schemes on real flights.

Accurate silhouette segmentation using motion detection and graph cuts

Acquiring accurate silhouettes has many applications in computer vision. This is usually done through motion detection, or a simple background subtraction under highly controlled environments (i.e. chroma-key backgrounds). Lighting and contrast issues in typical outdoor or office environments make accurate segmentation very difficult in these scenes. In this paper, gradients are used in conjunction with intensity and colour to provide a robust segmentation of motion, after which graph cuts are utilised to refine the segmentation. The results presented using the ETISEO database demonstrate that an improved segmentation is achieved through the combined use of motion detection and graph cuts, particularly in complex scenes.