Alistair Sutherland

School of Computing, Dublin City University, Dublin, Ireland

Phone: +353 - 1 - 7005511 Fax: 7005442

e-mail: alistair.sutherland at dcu.ie

Room: L1.04

Research Interests

Computer Vision, Human Motion Tracking, 3D Vision, Shape Recognition, Public Speaking, Positive Computing, Multi-modal Interaction

Current Research Students

Dexmont Pena

Dexmont is working on autonomous vehicle control using computer vision. He has developed techniques to compute a depth map of the scene in front of a vehicle. His algorithm processes the video streams from a pair of stereo-cameras. It uses image processing techniques based on the Census Transform and the Complete Rank Transform to match points across the two streams. The algorithm is computationally efficient because it computes matches only on edges and so computes a sparse depth map. He is currently working on extending this algorithm for obstacle detection.

Marlon Oliveira

Marlon is working on hand-shape recognition for mobile applications. He is trying to identify the hand-shapes from Irish Sign Language in images. He is using techniques based on Principal Component Analysis (PCA), which represents shapes as points in a low-dimensional space. Marlon is using multi-stage version of PCA, which applies PCA to subsets of points in an existing PCA space. This allows him to create a hierarchical search algorithm. He is currently developing techniques to interpolate between points in a PCA space and to interpolate between the spaces themselves.

Fiona Dermody

Fiona is working on a Multimodal Positive Computing System for Public Speaking with Real-Time Feedback

The system detects body pose, facial expressions and voice using the Microsoft Kinect. Visual and text feedback is displayed in real time to the user using a video panel, icon panel and text feedback panel. Real time feedback is displayed on gaze direction, body pose and gesture, vocal tonality, vocal dysfluencies and speaking rate.

The development of the system encapsulates inter-disciplinarity through its application of the principles of affective computing, psychology and the humanities, specifically the art of rhetoric.  The ultimate goal of this multifaceted approach is to increase user wellbeing through the informed utilisation of technology with inputs from humanities and the social sciences.

Initial user feedback on the prototype from a pilot study has been positive. In future work we will explore ways of using multi-modal feedback to reduce the cognitive load on the user.

Positive Computing is an appropriate, interdisciplinary framework, within which to develop a digital system for reducing public speaking anxiety, because the objectives of this research align with the objectives of positive computing.  Public speaking anxiety has a significant impact on a personís wellbeing. In particular it has been found to impact on success in academia, in society and in business. A system such as this will enable users to maximize their potential in these areas. By providing users with reflective feedback on their speaking performance, it can help them to become aware of how the mechanics of their speaking could be perceived. This heightened awareness could enable a user to reduce their speaking anxiety thereby potentially increasing their wellbeing.

Fionaís Publications

3rd and 4th year projects