|
Nicolas Pugeault, Dr. Rer. Nat.School of Computing Science,University of Glasgow, Glasgow G12 8RZ.
|
https://orcid.org/0000-0002-3455-6280I have now moved to the University of Glasgow (here)
Highlights and news
- We have a new paper accepted for Face and Gestures'2020: "Real-time Facial Expression Recognition "In The Wild" by Disentangling 3D Expression from Identity" (web).
- Where do we look when describing visual scenes? We have a new paper out investigating visual attention while describing images, "Human Attention in Image Captioning: Dataset and Analysis", accepted for IEEE ICCV'2019 (arXiv)(dataset)
- We have two articles accepted for IEEE ICDL-Epirob, this year in Oslo: "Training-ValueNet: Data Driven Label Noise Cleaning on Weakly-Supervised Web Images" (pdf) and "Contextual Relabelling of Detected Objects" (arXiv)
- New research out: our paper has been accepted for IEEE CVPR: "Understanding and Visualizing Deep Visual Saliency Models". (arXiv)
- New research: How to improve trust region based reinforcement learning by reusing data from several consecutive policies? Our working paper (arXiv) (code)
- Our paper "Learning Context on a Humanoid Robot Using Incremental Latent Dirichlet Allocation", IEEE TCDS, 2016 has been awarded the 2019 IEEE TCDS Outstanding Paper Award! (PDF)
- How much of driving is pre-attentive? Quite a lot, it appears. This is the focus of new research, in collaboration with Richard Bowden. In there, we demonstrate that up to 80% of steering and braking actions can be explained by a pre-attentive model, and that the driver's steering can be predicted accurately up to a full second before he turns the wheel. (pdf)(doi)
Research Interests
My core research interests are cognitive vision and machine learning. More specifically, I am interested in how internal representations of visual information can emerge from interaction with an agent's environment. This involves aspects such as feature extraction and selection, perceptual grouping of features, accumulation and disambiguation of visual information over time, discovering objects' shape from experience, and more. Moreover, I have a keen interest in biologically plausible functional models of human vision.
Brief Biography
After finishing my Engineer studies at the ESIEA Paris, I moved to UK and did an MSc in Computational Intelligence at the University of Plymouth, before moving to Stirling to study with Norbert Krueger and Florentin Woergoetter on the European project ECOVISION. I obtained my PhD from the University of Goettingen in 2008. From September 2007 to September 2009, I held a half-time Assistant Professor position at the University of Southern Denmark, and a half-time Research Associate position at the University of Edinburgh, collaborating with Prof. Norbert Krueger on the two EU projects DrivSco and PACOPLUS. I have been employed as a Research Fellow at CVSSP since 2009, working with Richard Bowden on cognitive vision and machine learning research (I was involved with the projects DIPLECS and Dicta-Sign), before starting as a lecturer in January 2013.
From January 2016 to May 2020, I was with the Computer Science department at the University of Exeter, before Joining the University of Glasgow in June 2020.