Caroline Baillard

Principal Scientist

Immersive Lab Researchers

Caroline is a principal scientist and Technical Area leader of the team “Mixed Reality”. Her main research interests are computer vision and scene analysis. She graduated from Telecom Paris Tech in 1994 and received a Ph.D. degree in Signal and Image processing from Telecom Paris Tech in 1997. After 2 years in Oxford University (UK), and 1 year at INRIA Rennes, she led the R&D team of SIRADEL Map Data department for more than 10 years, a company specialized in the production of virtual 3D city models. She joined Technicolor R&I in 2011 as a principal scientist and project leader. Since then she has been leading teams up to 30 people, and she has been involved in several research projects focusing on facial motion capture and retargeting, depth-based video editing, or Augmented Reality for home entertainment. She is author or co-author of more than 30 papers.

Research interests:

  • Augmented and Mixed Reality,
  • 3D reconstruction and modelling,
  • Point cloud analysis,
  • Photometric analysis,
  • Camera pose estimation,
  • Data registration,
  • Image and video processing
  • Digital cartography,
  • Face analysis & animation.


  • Mixed Reality Immersive TV:
    Towards personalized MR experiences at home The technical goal of the project is to build innovative technology portfolio for improving the blending quality in Mixed Reality (MR): solutions for camera pose tracking, scene 3D modeling, photometric analysis… The business potential of Mixed Reality for home entertainment is illustrated through a new muti-device experience called “MR extended TV”, where MR is used to extend and enhance a standard TV content with a new type of personalized and interactive immersive content.
  • Depth-based video editing
    Design and development of an acquisition & modeling system for mobile devices enabling enhanced 3D-based video editing solutions.
  • Facial motion capture and retargeting
    Characterization of the performance of an actor to reproduce it onto a CG character. The target application is to automate non-creative animation tasks during video game or animation films production.