The paper "Probabilistic Articulated Real-Time Tracking for Robot Manipulation" by Cristina Garcia Cifuentes, Jan Issac, Manuel Wüthrich, Stefan Schaal and Jeannette Bohg was finalist for the Best Robotic Vision paper at the 2017 IEEE/RAS International Conference on Robotics and Automation.
The paper proposes a probabilistic filtering method which fuses joint measurements with depth images to yield a precise, real-time estimate of the end-effector pose in the camera frame. This method in combination with our previous work on object tracking endows a robust with robust hand-eye coordination.
Along with the paper, we release open-source code on Bayesian Object Tracking and data sets annotated with ground truth on rigid or articulated object tracking.
Our goal is to understand the principles of Perception, Action and Learning in autonomous systems that successfully interact with complex environments and to use this understanding to design future systems