Tight Learned Inertial Odometry (TLIO) - Staircase

TLIO is an IMU-only Bayesian state estimation approach with a learned motion prior. The video shows results comparing TLIO to Visual-Inertial Odometry (VIO) and network-only velocity integration (RONIN) methods on pedestrian datasets.
Paper: Wenxin Liu, David Caruso, Eddy Ilg, Jing Dong, Anastasios I. Mourikis, Kostas Daniilidis, Vijay Kumar, Jakob Engel, "TLIO: Tight Learned Inertial Odometry", accepted at IROS 2020.
Webpage: cathias.github... - for all updated info about the paper, code and results.
Video Design: David Caruso
More videos:
Symposium: • Tight Learned Inertial...
Pool: • Tight Learned Inertial...