Elektor is excited to announce that for Elektor Engineering Insights (EEI) #57, they have secured OpenMV President and co-founder Kwabena Agyeman as the guest speaker. The session, hosted by Elektor, will delve into the realm of practical computer vision specifically tailored for microcontrollers and other hardware with constraints.
Unlike the conventional approach of running computer vision on a Linux system, this discussion will focus on treating vision as a sensor integrated within an embedded system. The emphasis will be on a firmware stack and tooling designed for real-world deployments, highlighting essential considerations such as hardware choices, image pipelines, performance metrics, power efficiency, and debugging processes in non-desktop environments.
If you are interested, you can sign up for the livestream or catch the recording later. Additionally, background materials including project documentation and the source repository will be provided for further exploration.
The upcoming conversation will address the practical applications of camera-and-MCU platforms compared to Single Board Computers (SBCs) and supplementary accelerators. It will also shed light on how the development workflow facilitates rapid iteration and the reality of "Edge AI" claims when deploying models within strict memory and compute constraints. Expect a deep dive into OpenMV embedded vision without the usual marketing hype.
During the live show, there will be an exciting giveaway of five Arduino Pro Portenta Vision Shield (Ethernet) boards. This event aligns with the recent focus of Elektor Engineering Insights sessions, as highlighted by eeNews Europe, which prioritize practical engineering discussions over product demonstrations.
Mark your calendars for Wednesday, 28 January 2026, at 16:00 CET (15:00 UTC) to join the live online session featuring Kwabena Agyeman from OpenMV. The format will include a Q&A session, providing attendees with the opportunity to engage directly with the speaker and delve deeper into the world of OpenMV embedded vision.