Research
My current research interests include 2D/3D image and video compression, computer vision, computational imaging, and frugal AI for multimedia signal processing. I am particularly interested in developing compact and effective representations of 2D and 3D visual data for efficient transmission and storage, as well as for solving a variety of inverse problems. I am also interested in bridging conventional visual compression frameworks with novel learning-based approaches while maintaining low computational complexity.
You can find my publications on my Google Scholar profile.
Current PhD students
- Théophile BLARD, 2023-2026, CIFRE with Orange Innovation
Neural video coding with adaptation and competition - Oumaima BADI, 2024-2027
Photorealistic human motion and style transfer using generative AI for improving sports training - Matthieu GENDRIN, 2024-2027, CIFRE with Orange Innovation
Modeling and compression of videos using 3D Gaussians - Massimo EYNARD, 2025-2028, CIFRE with Ateme
Video compression for high-quality immersive applications - Florent MUSSARD, 2026-2029
Frugal and embedded data compression for machine vision based on genetic reinforcement learning
Former PhD students
- Jinglei SHI (now lecturer in Nankai University, China), 2017-2020
Learning-based depth estimation from light field and view synthesis - Idi BOUBACAR SANI, 2023-2026
Deep learning methods for detecting ground movement - Pierre JAFFUER, 2025-2026, CIFRE with Orange Innovation
Low-complexity neural video coding
