RESEARCH AND ENGINEERING EXPERIENCE
After studies in computer science and applied mathematics (ENSIMAG 1987), and an experience in Computer Science teaching, I have been a researcher in Indexing at Thomson and then Technicolor since October 1990.
I have worked in multiple domains, mainly Image Indexing, Signal Processing for Audio, Text, User Profiles, and Pattern Recognition on images.
I moved in 2015 to the Data Processing Lab, to apply my strong experience in Machine Learning to the emerging fields of User Analytics and IOT. My ambition is to develop original and innovative methods for the decision part of IOT systems, for applications based on habits monitoring, actions detection or recognition, or non-intrusive person recognition.
My general expertise is about algorithmic methods for data analysis. Using a strong background in mathematics I had the opportunity to apply and to develop new methods in related fields such as:
- Pattern recognition, classification methods, unsupervised classification, data analysis,
- Optimization methods, numerical analysis,
- Signal processing for image and audio,
- Image retrieval,
- Recommendation systems, user profiling.
- Sequences, Items and Latent Links: Recommendation with Consumed Item Packs, R. Guerraoui, E. Le Merrer, R. Patra, JR Vigouroux, submitted to ACM Transactions on Internet Technology (January 2018).
- Time-Aware User Identification with Topic Models, Clément Lesaege, François Schnitzler, Anne Lambert and Jean-Ronan Vigouroux, IEEE ICDM, December 2016.
- Probabilistic Geometrical Verification, JR. Vigouroux, F. Lefèbvre, C. Thébault, Submitted to ACM Multimedia 2015.
- Facial Landmarks Localization Estimation by Cascaded Boosted Regression, L. Chevallier, JR. Vigouroux, A. Goguey, A. Ozerov, Visapp 2013. Selected for publication in Computer Vision, Imaging and Computer Graphics – Theory and Applications, Springer, 2014.
- On Evaluating Face Tracks in Movies, A. Ozerov, JR. Vigouroux, L. Chevallier, P. Pérez, ICIP 2013.
- Automatic categorization of short textual metadata of multimedia database in replacement of a rule-based classification system, Louis Chevallier & Jean-Ronan Vigouroux (Thomson Multimedia, FR) Ted Diamond & Eric Rehm (Singingfish, US), SIGIR 2002 Operational Text Classification Systems Workshop.
Twenty patents granted today, mainly in the fields of image and signal (music indexing), user interfaces, video compression.