Supervision: Johann Paratte

Project type: Semester project (master) Master thesis


Idea : provide a scheme to automatically infer the 3D model of the static background in a scene from a single video sequence including dynamic objects.

Objectives :

  • compute a background-foreground segmentation from the deformation of a keypoint stream in a sequence
  • reconstruct a 3D model of the background

Skills required :

  • Programming (C++)
  • Notions of Computer Vision