Overview Before you begin Define the problem and the model Set up environment Build basic camera application Transform pre-trained Caffe model Build image recognition application CMSIS-NN Deploy transformed model on Cortex-M MCU Troubleshooting Next steps
Single Page Download PDF

Overview

This guide shows you how to perform real-time image recognition on a low-power Arm Cortex-M7 processor, using Arm’s CMSIS-NN library. The Cortex-M7 processor is found in a range of solutions from a variety of microcontroller vendors.

Watch a video demo of the steps in this guide:


Increased compute performance in the smallest Cortex-M based devices is enabling more data processing to move to the edge. Machine learning (ML) no longer needs to take place in the cloud or only on advanced application processors. Now, you can take advantage of the benefits afforded by edge computing, such as reduced bandwidth and time-to-decision, increased privacy and reliability, all without relying on an internet connection. 

ML at the edge offers new possibilities for new real-time decision-making applications, from voice recognition for smart speakers to facial detection for surveillance cameras.

Next