Stéphane Lathuilière
Also known as Stephane Lathuiliere
Stéphane Lathuilière works on computer vision and multimodal learning. He was an associate professor at Télécom Paris from 2019 to 2024. He co-authored a 2022 manuscript on generating head motion from a single reference pose with an autoregressive generative adversarial network. He joined the RobotLearn team in 2024 through an Inria Starting Faculty Position. He works on adapting foundation models and has taught at UGA and Ensimag since 2025.
- Now at
- Inria and Université Grenoble Alpes
- Works
- 108
- Citations
- 2,701
- h-index
- 24
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2019–2024
AI and vision at Télécom Paris
Lathuilière was an associate professor at Télécom Paris from 2019 to 2024. He taught in AI and computer vision programmes.
-
2022
Generating head motion
Lathuilière co-authored a manuscript with Louis Airale, Xavier Alameda-Pineda, and Dominique Vaufreydaz, submitted in November 2022. It presents a generative adversarial network for head motion from a reference pose. Incremental outputs and a multi-scale discriminator address smooth trajectories and varied motion frequencies.
-
2024
Joining RobotLearn
He joined the RobotLearn team in 2024 through an Inria Starting Faculty Position.
-
2025
Teaching deep learning and vision
He has taught at UGA and Ensimag-Grenoble INP since 2025, focused on deep learning and computer vision. He also continues to teach in some Paris programmes. He also works as a RobotLearn research scientist.
Sources and credits
Photo: stelat.eu; creator not stated. License: Not stated by source. Source: stelat.eu. Cropped to a square and resized. Ask us to remove or replace this photo.