no code implementations • NeurIPS 2020 • Yuriy Biktairov, Maxim Stebelev, Irina Rudenko, Oleh Shliazhko, Boris Yangel
To be useful within a real-time autonomous driving pipeline, a motion prediction system must provide efficient ways to describe and quantify this uncertainty, such as computing posterior modes and their probabilities or estimating density at the point corresponding to a given trajectory.