معرفی
Christophe Kervazo is an Assistant Professor (Maître de Conférences) at Télécom Paris, France, affiliated with the IMAGES group under the Image, Data, Signal (IDS) department. His research focuses on sparse blind source separation, nonnegative matrix factorization, hyperspectral imaging, and optimization techniques for remote sensing and biomedical applications.
- Education: Engineering degree from Supélec (2015), Master of Science from Georgia Institute of Technology (2016), PhD in Signal and Image Processing from Université Paris Saclay (2019).
His work spans deep learning for inverse problems (including deep unrolling techniques), remote sensing (hyperspectral imaging and SAR), and uncertainty quantification. Recent publications address synthetic data training for medical imaging, distributed sparse BSS, and nonlinear component separation. Collaborators include institutions like CEA Saclay, Université de Mons, and ONERA.
Current students include PhD candidates working on topics such as digital breast tomosynthesis, hyperspectral unmixing, and SAR image reconstruction. Former students and interns have contributed to projects involving plug-and-play methods, unrolling algorithms, and implicit regularization.
He is involved in teaching and research projects, including collaborations with Airbus and ONERA on hyperspectral imaging and spectral band optimization. His lab, LTCI (Information Processing and Communication Laboratory), supports interdisciplinary work in signal and image processing.
