Date: Thursday, May 23
Start Time: 4:15 pm
End Time: 4:45 pm
In this talk, we share highlights from our adventures developing computer vision solutions for the rail transportation industry. We begin with an introduction to the types of machine perception problems encountered in this unique industry. We then share insights gained during our implementation of a perception system for rail auditing. In particular, we discuss the challenges we faced in implementing multiple optimized CNNs in a constrained compute system, and how we addressed these challenges. We also explain trade-offs associated with using classical homographic techniques in combination with neural networks.
