Location. Courtyard, Alamo ABC

Date. Friday, December 11

Time. 8:30-12:30

Overview:

Human body models play an important role in human pose detection, tracking, and activity analysis. Particularly today, such models are useful for training machine learning methods.

Previously, highly accurate models were either not available or not practical for computer vision. In this tutorial we take participants through the recent history and the concrete steps needed to build a high quality model.

This is a "how to" tutorial that will cover the steps of

  • data capture
  • mesh alignment
  • choice of shape representation
  • dealing with varying body shape
  • modeling articulation
  • capturing and modeling soft-tissue motions
  • fitting models to data
  • applications in medicine, psychology, VR, animation, fashion

At the end attendees will walk away with a 3D body model that they can readily use for their own research purposes.

Who is it for?

The tutorial is designed for vision students and researchers. It does not assume strong graphics experience. If you need a 3D body model for training, animation, pose estimation, etc., you will get the tools you need here to go further.

Instructors

The tutorial is taught by a team from the Max Planck Institute for Intelligent Systems, who have spent years developing 3D human body models for vision, graphics and learning. They will provide both theoretical background and practical tools.