Mathis is a Research Scientist at NVIDIA working on text-to-3D human motion synthesis. He received his ELLIS Ph.D. from École des Ponts ParisTech (ENPC) as part of the IMAGINE computer vision team, in close collaboration with the Perceiving Systems department at the Max Planck Institute for Intelligent Systems (MPI-IS). His research focuses on generating realistic, diverse, and controllable human motion from labels or natural language, and on learning joint text-motion latent spaces that scale beyond expensive motion capture data.