Distracted driving is one of the leading causes of auto accidents, according to the National Highway Traffic Safety Administration (NHTSA). This talk will demonstrate the use of Artificial Intelligence to analyze driver images and identify distracted driving behavior autonomously. Two deep learning models were created using videos of drivers from a 3D image sensor and a 2D web camera. An ensemble of the models was used to classify the action of the driver. I will discuss the methodology, results and suggested areas of future work to improve driver safety. Key Takeaways: 1. Self-Driving cars and distracted driver detection can together provide a great driving experience. 2. We were able to identify distracted driving behavior with high accuracy, in real-time. 3. Future work - this can be integrated with self-driving car technology for a safe, fun driving experience.
Priya Sundararaman is a Principal Data Scientist at State Farm. Priya has an undergraduate engineering degree in Computer Science and masters in Predictive Analytics with 16 years of industry experience. She is a pragmatic data scientist who believes that we are already in the midst of the fourth industrial revolution, with AI being a key enabler, permeating all aspects of business. At State Farm, she intends to make her contribution by using machine learning to solve hard business problems for demonstrable, measurable, and sustainable ROI.