Gian Marco Iodice, Software Engineer at ARM, demonstrates the company's latest embedded vision technologies and products at the May 2017 Embedded Vision Summit. Specifically, Iodice presents a demonstration developed by partner company Au-Zone Technologies. Using a deep neural network running on a standard embedded ARM processor (the Rockchip RK-3288), the demonstration classifies driver activity in real time with an accuracy greater than 95%. The solution was developed entirely using Au-Zone’s DeepView toolkit, with an inference engine accelerated on the Mali-T760 GPU and ARM CPUs.
