With the popularity and success of our first Introduction to Embedded Machine Learning course, we decided to launch another! We listened to feedback from students, engineers, and industry leaders about which areas in tinyML were most interesting and useful. One topic stood above the rest: vision. Shawn Hymel returns as the main instructor, and we teamed up with OpenMV, Seeed Studio, and the tinyML Foundation to create a new course: Computer Vision with Embedded Machine Learning.
The course covers important concepts in computer vision, including how digital images are constructed, stored, and manipulated. You will learn how to create your own dataset from captured images and use Edge Impulse to train a neural network to classify these images. From there, you will see how convolutional neural networks (CNNs) operate and how they can be used to create robust machine learning models for vision applications. Finally, you will have the chance to train your own object detection system, which allows you to identify multiple objects of interest in images and videos.
You will see how to use the Edge Impulse tool to train your own models, and you will have the opportunity to deploy trained models to a few different embedded systems. At launch, the course provides instructions for deploying these models to an OpenMV H7 Camera as well as a Raspberry Pi 4.
The course is a follow-on to Introduction to Embedded Machine Learning, as it assumes you have some familiarity with machine learning concepts, including collecting data, feature extraction, and training. As a result, we recommend taking the Introduction to Embedded Machine Learning course first if you do not have any machine learning experience.
The Computer Vision with Embedded Machine Learning course is free for everyone with an optional professional certificate available. Let us know what other courses you would like to see, as embedded machine learning is a fast growing field!