Making Cozmo See
In this session, Andrew Stein will present the vision capabilities of Cozmo, Anki's latest consumer robotics product. Cozmo is a sophisticated entertainment robot focused on personality, interactivity, and game play. It was one of the hottest toys of 2016, earning a spot amongst the top two best-selling premium toys of the holiday season. In addition to covering high-level architecture and functionality, Stein will discuss some of the challenges of building a capable vision system at a scale and price point for the consumer market.
Andrew Stein is Lead Computer Vision Engineer at Anki, where he began his work on the Cozmo project four years ago as the team’s first member. He has contributed to several core systems of the product, including vision, cube manipulation, animation streaming, localization, high-level behaviors, and low-level actions. Prior to joining Anki, Stein worked on intrinsic imaging and video at Tandent Vision Science, stereo vision at NASA’s Jet Propulsion Laboratory, hyper-spectral image analysis at ChemImage, and structure from motion at Intel. He earned his Ph.D. from the Robotics Institute at Carnegie Mellon University, and his Bachelor’s and Master’s degrees in Electrical and Computer Engineering from the Georgia Institute of Technology. When he’s not bringing cute, friendly robots to life, Stein generally can be found on a nearby tennis court.