Google Maps provides useful directions, real-time traffic and information on businesses to millions of people every day. This information has to constantly mirror the ever-changing world to provide the best experience for our users. Google StreetView cars collect millions of images daily. We leverage large scale machine learning and computer vision algorithms to extract semantic information from Street View imagery to improve Google Maps. This talk will focus on some of core problems solved, lessons learned, and interesting research challenges that lie ahead.
Dr. Rajesh Parekh is Engineering Director at Google where he leads a talented team focusing on algorithmic data curation for Google’s Geo products. His team applies various Machine Learning and Computer Vision techniques to model the real-world. He is passionate about solving challenging problems that deliver tremendous user impact.
Prior to Google, Dr. Parekh led analytics for Facebook’s Video and Applied Machine Learning initiatives. Before Facebook, Dr. Parekh was the Vice President of Data Science at Groupon where he built products for personalization, sales automation, and marketing optimization. He also worked at Yahoo Labs building display advertising targeting products, at Blue Martini Software developing data mining products for e-commerce, and at Allstate solving insurance problems like cross-sell, retention, and fraud.
Dr. Parekh earned his Ph.D. in Computer Science focusing on Artificial Intelligence from Iowa State University. He has published over 30 research papers and actively participates in the machine learning and data science community.