Dr Samuel Pachoud, PhD
Vision Group, Computer Science
Tel: +44 020 7882 8018
IntroductionSamuel is currently an advisory consutlant at Ernst & Young in the performance improvement division. He previously did a PhD in Queen Mary, University of London under the supervision of Prof. Shaogang Gong and Prof. Andrea Cavallaro. Research TopicHis research interest focused on interdisciplinarity between Electronic Engineering and Computer Science. His thesis addresses the problem of extraction and combination of audio and visual information for increased robustness and accuracy in recognition, especially in noisy environments. It focuses on three problems of an audio-visual speech and emotion recognition system: feature representation, feature selection and feature fusion. Short BioSamuel was born in Estavayer-le-Lac, Switzerland on April 26, 1982.
He always lived
in Payerne, Switzerland until he came in London in September 2006. Samuel
received the
MSc degree in Electrical and Electronic Engineering from the Swiss
Federal
Institute of Technology, Lausanne, Switzerland in April 2006. In 2005,
He visited
for 6 months the Multimedia and Vision group at the Queen Mary
University of
London, working as a guest research student in the area of clustering
and
behaviours analysis. The purpose was to summarise the dynamics
behaviour of
multiple objects by clustering object trajectories. Research InterestsSamuel's main research interests were clustering method, speech processing and recognition, lip-reading, audio-visual perception, face modelling and surveillance + security applications. PublicationsJournal papersAudio-Visual Fusion using Kernel Canonical Correlation Analysis for Noisy Speech RecognitionS. Pachoud, S. Gong, A. Cavallaro IEEE Transactions on Multimedia, under submission Category-sensitive audio-visual feature selection for emotion recognition S. Pachoud, A. Cavallaro, S. Gong IEEE Transactions on Multimedia, under submission Conference papersSpace-time audio-visual speech recognition with multiple multi-class probabilistic Support Vector Machines [pdf]S. Pachoud, S. Gong, A. Cavallaro International Conference on Auditory-Visual Speech Processing (AVSP), Norwich (UK), September 10-13, 2009 Grouping motion trajectories [pdf] S. Pachoud, E. Maggio, A. Cavallaro IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Taipei, Taiwan, April 19-24, 2009 Video augmentation for improving audio speech recognition under noise [pdf] S. Pachoud, S. Gong, A. Cavallaro British Machine Vision Conference (BMVC), Leeds (UK), September 1-4, 2008 Macro-cuboids based probabilistic matching for lip-reading digits [pdf] S. Pachoud, S. Gong, A. Cavallaro IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), Anchorage, Alaska (USA), June 24-26, 2008 MiscellaneousAutomatic macro-block based feature selection for lip-reading [pdf]S. Pachoud, S. Gong, A. Cavallaro BMVA Inaugural Student Papers Meeting, London (UK), 28 March, 2007 Master thesisS.Pachoud, "Clustering and evaluation of object trajectories" [pdf]M.Sc. Thesis, Swiss Federal Institute of Switzerland, April 2006 Advisor: Prof. Dr. J-P Thiran and Dr. A. Cavallaro. Assistant: Dr. E. Maggio Activities
AwardsFull Research Studentship, Queen Mary, University of London, 2006-2009Overseas Research Students Awards Scheme (ORSAS) award, 2007-2009 Teaching Assistant
|