In this episode, we’re joined by Hanbyul Joo, a PhD student in the Robotics Institute at Carnegie Mellon University.
Subscribe: iTunes / Google Play / Spotify / RSS
Han, who is on track to complete his thesis at the end of the year, is working on what is called the “Panoptic Studio,” a multi-dimension motion capture studio with over 500 camera sensors that are used to capture human body behavior and body language. While robotic and other artificially intelligent systems can interact with humans, Han’s work focuses on understanding how humans interact and behave so that we can teach AI-based systems to react to humans more naturally. In our conversation, we discuss his CVPR best student paper award winner “Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodies.” Han also shares a complete overview of the Panoptic studio, and we dig into the creation and performance of the models, and much more.
TWIML Online Meetup Update!
A few meetup related announcements:
- Our next TWIML Online Meetup is coming up on September 19th, and will feature David Clement presenting the paper “DeepMimic: Example-Guided Deep Reinforcement Learning of Physics-Based Character Skills.” This should be a fun one and I encourage you to join us.
Our second session of the Fast.ai study group is starting soon. If you have some coding experience and would like to learn state of the art deep learning, this is a great way to do it.
Join both of these great groups by signing up at twimlai.com/meetup!
Mentioned in the Interview
- Panoptic Studio Database
- OpenPose Library
- CVPR Best Student Paper Award Winner: Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodies
- TWIML Presents: Series page
- TWIML Events Page
- TWIML Meetup
- TWIML Newsletter
“More On That Later” by Lee Rosevere licensed under CC By 4.0