Deep Dive into TensorFlow #5

Please note that the meetup's RSVP link (upper right corner) does not work and it will appear as waitlist.

Instead, please register HERE (http://bit.ly/2vdY7Ia)

Many thanks to Uber (http://eng.uber.com/)for hosting and sponsoring the meetup!

https://secure.meetupstatic.com/photos/event/3/c/e/3/600_463215587.jpeg

Agenda:

6:30 Doors open. Networking. Pizza and drinks

7:00 Talk #1

Abstract Summary of Talk: How machine learning and computer vision approaches applied to street-level and satellite imagery can be used to build maps.

Ben Kadlec: Ben is the engineering manager for the Uber Maps Information Extraction Team. Ben's team focuses on building maps from street-level and satellite imagery by leveraging machine learning and computer vision approaches. Before coming to Uber, Ben worked on Bing Maps at Microsoft. In his distant past, he co-founded TerraSpark Geosciences which was later acquired by CGG. Ben has a PhD in Computer Science from the University of Colorado.

7:40 Talk #2

Abstract Summary of Talk: How computer vision can be used to optimally process 10s of millions of images simultaneously resulting in a highly accurate road network that could eventually replace the need for GPS to position imagery.

Aaron Rogan: Aaron is the engineering manager for the Uber Maps Camera & Sensor’s team. This team is focused on building next generation image collection platforms and exploiting that imagery to achieve the highest accurate geolocation of possible for the imagery. Prior to Uber Aaron was the Tech Lead for Information Extraction at Microsoft pattening multiple LiDAR processing techniques. He has a PhD in Physics from Washington State University where he wrote the LIGO’s initial coherent detection pipeline responsible for the first direct detection of gravitational waves in 2016.

8:20 Wrap-up