Ram Cherukuri, MathWorks
See how you can generate code from a trained deep neural network in MATLAB® for Arm® processors that support the Neon instruction set architecture (ISA) like the Arm Cortex®-A family. A demo shows pedestrian detection as an object detection application using deep learning on a Raspberry Pi™ 3 as the Arm target.
The generated code leverages the Compute Library, which is a collection of low-level software functions optimized for certain Arm architectures, targeting image processing, computer vision, and machine learning applications. The demo shows the neural network for pedestrian detection running on the Raspberry Pi 3 at 6 fps.