Maxime Ossonce
Professor and PhD student in Signal processing and Telecommunications under the supervision of L2S (Signals and Systems Laboratory, CentraleSupélec / Paris-Saclay)
ESME
Overview of research area
Image watermarking, Robustness of deep neural networks (out of distribtuion, misclassification detection).
Short biography
Professor at ESME since january 2016, I am currently fulfilling a PhD at LSS (Lignals and Systems Laboratory, CentraleSupelec/Paris-Saclay) on the study of robustness of deep learning models from the perspective of information theory.
Publications
- M. Ossonce, F. Alberge, and P. Duhamel, Joint classification and out-of-distribution detection based on structured latent space of variational auto-encoders, 2022 IEEE International Conference in Image Processing (ICIP), Bordeaux, France, October 2022
- M. Ossonce, C. Delpha, and P. Duhamel, Rotation and scale insensitive image watermarking, 2004 International Conference on Image Processing (ICIP 2004), Vol. 4, pp. 2611-2614, Singapore, October 2004.
- M. Ossonce, G. Le Guelvouit, C. Delpha, and P. Duhamel, Game theoretic resolution of the rotation/scale insensitive watermarking problem, SPIE International Symposium on Electronic Imaging Science and Technology, San José, California, United States, January 2006.