Recently, new embedded AI solutions are being developed more and more, but did you know that it is possible to implement AI even with microcomputers with limited memory resources? Therefore, this time, e-Force will introduce a small camera that can be obtained at a low price and an AI implementation method using RTOS. Conventionally, when performing object recognition using AI using camera images, processing was performed on the server side via the network, so communication costs and real-time performance were concerns. Edge AI is the means to solve this. In this seminar, we will implement AI on a microcomputer (STM32) and RTOS (μC3) with low memory resources and explain object recognition by edge AI using camera images. In the future, if you are considering edge AI using a cheap microcomputer, please take a look. ・How to acquire data from the camera and display it on the display ・Method of acquiring data from the camera and determining whether or not a person is captured by AI