Aline Roumy
My research interest is in compression for image/video. I design mathematically founded algorithms for visual data processing such as analysis, storage, access, and transmission.
My expertise lies in signal and image processing, coding theory, information theory and machine learning.
I've been a member of INRIA since 2001. Before that I was at Princeton University and I got my PhD from the University of Cergy-Pontoise in collaboration with Thales. I've also visited Eurecom and UC Berkeley.
I served as a member of the French National University Council (CNU 61) (2011-2015).
I served as a member of the Inria Evaluation Committee (2019-2023).
I was an Associate Editor for the IEEE Transactions on Image Processing (2018-2023), and the Annals of Telecommunications (2016-2020).
I am currently a Senior Associate Editor for the IEEE Transactions on Image Processing journal (since Nov. 2023).
News
- [May 24] Workshop of the PEPR MoleculArXiv.
- [Jun 22] Navid Mahmoudian Bidgoli received the i-PhD award from BPI for his Anax Startup Project, with Simon Evain (see page 41).
- [Jan 21] Navid Mahmoudian Bidgoli received an Inria Startup Studio grant for his ANAX project.
Awards
- Top 3%" of the accepted paper at ICASSP 2023 for the paper ``Joint compression and demosaicking for satellite images'', by P. Bacchus and R. Fraisse and A. Roumy and C. Guillemot. (PDF)
- 2023 Outstanding Editorial Board Member Award for IEEE Transactions on Image Processing
- Top 10%" paper at ICIP 2015 for the paper "Universal lossless coding with random user access: the cost of interactivity" by A. Roumy and T. Maugey. (PDF)
- Best Paper Award "Francesco Carassa" 2011 awarded by GTTI (Gruppo Italiano Telecomunicazioni e Teoria dell'Informazione) for the paper "Lossy compression of distributed sparse sources: a practical scheme" by G. Coluccia, E. Magli, A. Roumy and V. Toto-Zarasoa.
Research activities
Image and video coding for multimedia communications- Compression for storage on DNA..
- Compression for interactive video communication.
- Learning for video compression.
- Compressive sensing, for coding of correlated sparse sources.
- Error-resilient video coding, unequal erasure protection and bundle packet protection.
- Universal and ergodic distributed source coding.
- Coding for computing.
- Zero-error source coding (with decoder side information).
- Interactive compression.
- Inverse problems (super-resolution and inpainting).
Research projects
- Project CoMET: Coding for iMage rETrieval, project funded by the Labex CominLabs, Jan. 2025-Dec. 2025.
- PEPR MoleculArXiv on DNA storage, 2022-2029.
- Project CoLearn: Coding for Learning, project funded by the Labex CominLabs, Jan. 2021-Dec. 2025. Project website
- MediaKind Cifre contract on learned downscaler for video streaming, 2024-2027.
- MediaKind Cifre contract on new representation for multi-profile video compression, 2020-2023.
- Ateme Cifre contract on learning and video compression, Aug. 2020-Aug. 2023.
- 2 PhD contracts with Airbus on low light imaging, Sept. 2020-Oct. 2023.
- ENS Grant on combinatorial and strategic aspects in information theory, Oct. 2020-Oct. 2023.
- Inria Grant (ADT) for the technological development of an Interactive Coder for Omnidirectional Videos, Oct. 2020-Oct. 2022.
- PI of the project InterCom: Interactive video communication: project funded by the Labex CominLabs, Oct. 2016-Dec. 2020. Project website
- DGA Grant on deep learning image compression and classification, Oct. 2015- Sept.2018.
- Ericsson CIFRE projet on High Dynamic Range image/video compression, Oct. 2014- Sept.2017.
- 2 collaborations in the framework of the Joint research laboratory between INRIA and Alcatel Lucent Bell Labs, Jan. 2011-Jan. 2014.
- ANR ARSSO: Adaptable, Robust, Streaming SOlutions, Funded by ANR, Jun. 2010- Dec. 2014.
- FP7-IST NOE NEWCOM++. Network of Excellence in Wireless COMmunication, Funded by EC, Jan. 2008-Apr. 2011.
- Project CoLearn: Coding for Learning, project funded by the Labex CominLabs, Jan. 2021-Dec. 2025. Project website