Feel free to contact me in general if you have good grades and are interested in research in transportation (see my research interests). Due to large volume of sollicitation by email, make sure to make a well articulated case for yourself and provide relevant information, or you will be ignored. Although all teaching is done in French at École Polytechnique, you may write your thesis in English.
There is now a RSS feed to keep you updated of my work. Ensure your browser supports CSS to fully enjoy this page, or get one that does !
My research interests focus on intelligent transportation systems, road safety and information technology for transportation (data collection, processing, using machine learning techniques, and visualization). After completing my Ph.D. in France in 2005, supervised by Dr. Alain Grumbach and Dr. Sophie Midenet, I was a postdoctoral research associate in the Transportation Engineering group of the University of British Columbia (UBC) with Dr. Tarek Sayed. I am now an assistant professor in transportation at the École Polytechnique de Montréal.
I am interested in really intelligent transportation systems, i.e. systems that actually exhibit intelligent features such as adaptation to changing conditions with minimal supervising. Using video sensors and computer vision techniques, road users' trajectories can be extracted automatically. This rich microscopic data can then be interpreted automatically to understand road users' behaviour and analyze road safety (without waiting for accidents to happen). A robust probabilistic framework was developed for automated road safety analysis. I am also very interested in mobility in general, in particular vulnerable road users, cyclists and pedestrians (or active modes of transportation). Research on them has been typically limited, in particular in regard of their importance.
Massive amounts of data are now collected continuously in our highly connected world, from geo-location devices in common cell phones to video sensors. If this data can be automatically analyzed, there are great opportunities for a better understanding and optimization of transportation systems, from their management to user information. The following is a list of keywords that are relevant to my research interests:
Finally, I am a supporter of open science for many reasons, both from a philosophical and moral point of view, and from a practical point of view. From a philosophical and moral point of view, it is the right thing to do, especially for publicly funded research institutions and it allows reproducible research: why should you trust my claims if you cannot replicate my work? From a practical point of view, it is a better method (open source software is a better software development technique) and my research benefits from collaboration and sharing code and data with you, as you reference my research and release publicly your improvements in turn). Join the movement!
Co-investigator on Bruno Agard's NSERC collaborative research and development grant on "the development of competitive tools through the integration of geographical information systems and data mining"
Member of the Network for Research and Innovation in Road Safety funded by the "Fonds québécois de la recherche sur la nature et les technologies" (FQRNT)) lead by François Bellavance (HEC Montréal)
Detailed accident modeling using microscopic data, grant of the Research and Innovation, Directorate of Ecole Polytechnique, 15000 $, 2010.
Automated classification of accidents and interactions using microscopic data, grant for young professors of the "Fondation de Polytechnique", 15000$, 2009.
Disclaimer: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. In most cases, these works may not be re-posted without the explicit permission of the copyright holder.
N. Saunier. Incidence de la régulation d'un carrefour à feux sur le risque routier. Editions Universitaires Européennes, 2010. [amazon]
T. Sayed, N. Saunier, G. Lovegrove, and P. de Leur. Advances in proactive road safety planning. In Canadian Multidisciplinary Road Safety Conference, Niagara Falls, Ontario, June 2010. [pdf]
K. Ismail, T. Sayed, and N. Saunier. Automated safety analysis using video sensors: Technology and case studies. In Canadian Multidisciplinary Road Safety Conference, Niagara Falls, Ontario, June 2010. First prize of the Canadian Association of Road Safety Professionals/Insurance Bureau of Canada 2010 Student Paper Competition [pdf]
B. Y. S. Khanloo, F. Stefanus, M. Ranjbar, Z.-N. Li, N. Saunier, T. Sayed and G. Mori. Max-margin offline pedestrian tracking with multiple cues. In Seventh Canadian Conference on Computer and Robot Vision. IEEE, 2010. [bib] [pdf] [page]
N. Saunier, T. Sayed and K. Ismail. Large Scale Automated Analysis of Vehicle Interactions and Collisions. In Transportation Research Board Annual Meeting Compendium of Papers, Washington D.C., January 2010. 10-4059. Accepted for publication in Transportation Research Record: Journal of the Transportation Research Board. [pdf (typos removed from the version in the proceedings)]
K. Ismail, T. Sayed and N. Saunier. Camera Calibration for Urban Traffic Scenes: Practical Issues and a Robust Approach. In Transportation Research Board Annual Meeting Compendium of Papers, Washington D.C., January 2010. 10-2715. [pdf]
K. Ismail, T. Sayed and N. Saunier. Automated Analysis Of Pedestrian-vehicle Conflicts: A Context For Before-and-after Studies. In Transportation Research Board Annual Meeting Compendium of Papers, Washington D.C., January 2010. 10-3739. Accepted for publication in Transportation Research Record: Journal of the Transportation Research Board. Pedestrian committee outstanding paper award. [pdf]
2009
N. Saunier, K. Ismail and T. Sayed. Camera Calibration for Automated Road Safety Analysis using Video Sensors. Technical report, University of British Columbia, December 2009.
N. Saunier and T. Sayed. A prototype system for truck signal priority using video sensors. In Annual Conference of the Transportation Association of Canada, Vancouver, B.C., October 2009. [pdf]
K. Ismail, T. Sayed and N. Saunier. Automated pedestrian safety analysis using video data in the context of scramble phase intersections. In Annual Conference of the Transportation Association of Canada, Vancouver, B.C., October 2009. [pdf]
N. Saunier, T. Sayed and K. Ismail. An object assignment algorithm for tracking performance evaluation. In Eleventh IEEE International Workshop on Performance Evaluation of Tracking and Surveillance (PETS 2009), June 2009. [bib] [pdf (2 typos corrected)] [proceedings]
A. Tarko, G. Davis, N Saunier, T. Sayed and S. Washington. Surrogate Measures of Safety. White Paper, ANB20(3) Subcommittee on Surrogate Measures of Safety, ANB20 Committee on Safety Data Evaluation and Analysis, April 2009. [pdf]
N. Saunier, K.Ismail and T. Sayed. Automated road safety analysis: Video-based road user classification. Technical report, University of British Columbia, February 2009. [bib]
K. Ismail, T. Sayed, N. Saunier and C. Lim. Automated Analysis of Pedestrian-Vehicle Conflicts using Video Data. Transportation Research Record: Journal of the Transportation Research Board, 2140:44-54, 2009. [bib] [pdf] [TRR]
K. Ismail, T. Sayed and N. Saunier. Automated Collection Of Pedestrian Data Using Computer Vision Techniques. In Transportation Research Board Annual Meeting Compendium of Papers, Washington D.C., January 2009. Reference 09-1122. [bib] [pdf]
2008
N. Saunier, W. Kang, T. Sayed and C. Lim. Truck signal priority. Technical report, Bureau of Intelligent Transportation Systems and Freight Security, University of British Columbia, 2008. [bib]
N. Saunier and T. Sayed. Probabilistic Framework for the Automated Analysis of the Exposure to Road Collision. Transportation Research Record: Journal of the Transportation Research Board, 2083:96-104, 2008. [bib] [pdf] [page] [TRR]
2007
N. Saunier, T. Sayed and C. Lim. Probabilistic Collision Prediction for Vision-Based Automated Road Safety Analysis. 10th International IEEE Conference on Intelligent Transportation Systems, Seattle, October 2007. [bib] [pdf] [page] [IEEE]
N. Saunier and T. Sayed. Automated Road Safety Analysis Using Video Data. Transportation Research Record: Journal of the Transportation Research Board, 2019:57-64, 2007. [bib] [pdf] [TRR]
2006
N. Saunier and T. Sayed. Clustering Vehicle Trajectories with Hidden Markov Models. Application to Automated Traffic Safety Analysis. International Joint Conference on Neural Networks, IEEE. Vancouver, July 2006. [bib] [pdf] [page] [IEEE]
N. Saunier and T. Sayed. A Feature-Based Tracking Algorithm for Vehicles in Intersections. Third Canadian Conference on Computer and Robot Vision, IEEE. Québec, June 2006. [bib] [pdf] [page] [IEEE]
N. Saunier and T. Sayed. Automated Road Safety Analysis Using Video Sensors. Technical report, University of British Columbia, 2006. [bib]
2005
Ph.D. thesis (2005), entitled "Influence of traffic control in a signalized intersection on the risk of road users; Stream-based learning of safety indicators through data selection". [bib] [server archive]
N. Saunier, S. Midenet and A. Grumbach. Stream-based Learning through Data Selection in a Road Safety Application. STAIRS 2004, Proceedings of the Second Starting AI Researchers' Symposium, volume 109 of Frontiers in Artificial Intelligence and Applications, pages 107-117, Valencia, August 2004. [bib] [pdf]
N. Saunier, S. Midenet and A. Grumbach. Apprentissage incrémental par sélection de données dans un flux pour une application de sécurité routière. Conférence d'Apprentissage, pages 239-251. Montpellier, June 2004. [bib] [pdf]
N. Saunier, S. Midenet and A. Grumbach. Coopération entre système d'inférence et système de reconnaissance; détection de situations dangereuses dans un carrefour à feux. Génie Logiciel, (71):20-24, December 2004. [bib] [pdf]
2003
N. Saunier, S. Midenet and A. Grumbach. Automatic detection of vehicle interactions in a signalized intersection. 16th ICTCT Workshop, Soesterberg, October 2003. [bib] [pdf, from ICTCT]
2000-2001
N. Saunier. Interaction Multimodale et Monde Virtuel, Rapport de Projet de 3ème année, ENST, Paris, July 2001. [pdf]
N. Saunier. Automatic adaptation of learning parameters in reinforcement learning, Internship Report, ATR International, Kyoto, December 2000. [ps.gz]
How Many Ways to Crash, Invited Presentation in the Session on Developments in Alternative Models for Road Safety, Transportation Research Board Annual Meeting, Washington, D.C., January 13th, 2010. [pdf]
Large Scale Automated Analysis of Vehicle Interactions and Collisions, Session on Varied Approaches to Road Safety Assessment, Transportation Research Board Annual Meeting, Washington, D.C., January 12th, 2010. [pdf]
Presentations on Road Safety Analysis (with Kareem El-Basyouny and Karim Ismail) and Video-based Transportation Data Collection (with Karim Ismail) at a joint Transportation Symposium of the University of British Columbia and University of Washington, Seattle, April 3rd 2009.
A Probabilistic Framework for the Automated Analysis of the Exposure to Road Collision (Poster), Transportation Research Board Annual Meeting, Washington D.C., January 14th, 2008. [pdf]
The 'I' of ITS, Introductory Course on ITS, UBC CIVL 440 course, Vancouver, November 21st, 2007.
Probabilistic Collision Prediction for Vision-Based Automated Road Safety Analysis, 10th International IEEE Conference on Intelligent Transportation Systems, Seattle, October 3rd, 2007. [pdf]
Vision-based Road Safety Analysis, invited talk at the SFU Vision and Media Lab with professor Greg Mori, Vancouver, September 12th, 2007. [pdf]
Automated Road Safety Analysis Based on Video Sensors, Meeting with Christer Hydén and Ase Svensson, UBC, Vancouver, June 28th, 2007. [pdf]
Automated Road Safety Analysis Using Video Data, Translink, Vancouver, January 29th, 2007. [pdf]
Clustering Vehicle Trajectories with Hidden Markov Models. Application to Automated Traffic Safety Analysis, International Joint Conference on Neural Networks, Vancouver, July 20th, 2006. [pdf]
A Feature-Based Tracking Algorithm for Vehicles in Intersections (Poster), Third Canadian Conference on Computer and Robot Vision, Québec City, June 7th, 2006. [pdf]
Vehicle Trajectories Clustering with Dynamic Bayesian Networks for Traffic Safety Analysis, ITS Canada Conference, Whistler, June 5th, 2006. [pdf]
Clustering automatique de trajectoires avec des HMMs; Application à la sécurité du trafic routier, Séminaire Machine Learning ENST, Paris, January 9th, 2006. [pdf]
Automated Road Safety Analysis Using Video Sensors, UBC Transportation Seminar, Vancouver, September 15th, 2005. [pdf]
Stream-Based Learning through Data Selection in a Road Safety Application, STAIRS 2004, Valencia, June 24th, 2004. [pdf]
Apprentissage incrémental par sélection de données dans un flux pour une application de sécurité routière, Conférence Cap 2004, Montpelier, June 14th, 2004. [pdf]
Coopération entre système d'inférence et système de reconnaissance; Détection de situations dangereuses dans un carrefour à feux, Forum NTIC et Transports sur le Soft Computing & Transports, Paris, June 27th, 2004. [pdf]
Automatic detection of vehicle interactions in a signalized intersection, 16th Workshop of the International Cooperation on Theories and Concepts in Traffic safety, Soesterberg, October 30th, 2003. [pdf]
As stated in my research interests I support open science, i.e. sharing data and code. There is not much available for now, but I am working on it. ;-)
Open source code
An example of the use of KLT and OpenCV together. Even if OpenCV has some feature-tracking functions, the KLT implementation is simple to use and efficient. However, its file reading capabilities are limited to pgm image files, while OpenCV can read avi file. Here is the example with the corresponding Makefile modified from KLT (tested under Cygwin on Windows XP).
This piece of code is released under the GNU General Public License (February 2006).
The dataset used in the TRB 2010 paper "Large Scale Automated Analysis of Vehicle Interactions and Collisions" will be made available. Please send me an email if you are interested. You may subscribe to the RSS feed to know when it is made available. There is also a lot of relevant research links on my del.icio.us account.
