Brian HoltCentre for Vision, Speech and Signal Processing,
Faculty of Engineering & Physical Sciences,
University of Surrey,
Guildford GU2 7XH,
My core research interests are computer vision and machine learning. More specifically, I am interested in human pose estimation using discriminitive machine learning techniques with application to Sign Language Recognition. Human pose estimation is an exciting research field, with many new techniques proposed over the last few years alone.
After finishing my Engineering studies at the University of Stellenbosch, I moved to the UK to work at Bloomberg and Markit as a software developer before accepting a PhD position with Prof. Richard Bowden to work on human pose estimation for Sign Language Recognition on the EU project Dicta-Sign.
Holt, B. and Bowden, R. (2012).
Static Pose Estimation from Depth Images using Random Regression Forests and Hough Voting.
7th International Conference on Computer Vision Theory and Applications (VISAPP'12)
Best Student Paper.
Holt, B., Ong, E-J., Cooper, H., Bowden, R. (2011). Putting the pieces together: Connected Poselets for Human Pose Estimation. In 1st IEEE Workshop on Consumer Depth Cameras for Computer Vision
Cooper, H., Holt, B., and Bowden, R. (2011) Applications: Sign Language Recognition, In Looking at People: Automatic visual analysis of humans.