Kaouther Messaoud

Welcome! I am an Assistant Professor at Télécom Paris, Institut Polytechnique de Paris, Hi!Paris chair. My research focuses on modeling, predicting, and generating motion in complex visual environments. I study motion as a unifying abstraction that connects representation learning, multimodal reasoning, trajectory prediction, and generative modeling, with an emphasis on generalization and physical plausibility. My work spans motion forecasting for autonomous systems and, more recently, diffusion-based video generation for spatially and temporally coherent dynamic scene modeling.

Before joining Télécom Paris, I was a Postdoctoral Researcher at the VITA Lab, EPFL, working under the supervision of Prof. Alexandre Alahi. I earned my Ph.D. in 2021 from Sorbonne University, France, where I conducted research on Deep Learning for Motion Prediction at the ASTRA Lab, INRIA Paris.