Eliminating Topological Errors in Neural Network Rotation Estimation Using Self-selecting Ensembles
Event Type
Technical Paper
This session WILL be recorded.
Interest Areas
Research & Education
Registration Levels
Ultimate Supporter
Ultimate Attendee
Exhibitor Ultimate
TimeFriday, 13 August 20219am - 10am PDT
DescriptionWe explain why quaternions are a bad choice for rotation representation in neural networks, how symmetric shapes complicate the problem of rotation estimation, and how to deal with them.