Embedded Machine Learning: computer vision on a Raspberry Pi Pico microcontroller

TinyML is a fast-growing multidisciplinary technology where innovations in embedded hardware, software, and machine learning enable a new class of smart applications through on-device, near-sensor inference on ultra-low power embedded systems. This opens the door to new types of edge services and applications that do not depend on cloud processing, but are based on edge inference and autonomous reasoning. TinyML offers great advantages in terms of privacy, latency, energy efficiency and reliability. 

To demonstrate the possibilities of TinyML, Sirris has rolled out a computer vision application on a low-cost microcontroller with open-source tools: we were able to deploy a MobileNet on a Raspberry Pi Pico with TensorFlow Lite for Microcontrollers. The model processes RGB images with a resolution of 48 x 48 pixels and classifies them into three classes, with an inference time of 250 ms.

During the webinar we will elaborate on this image classification task. The following topics will be discussed:

• The hardware, microcontroller and camera
• The applied MobileNet architecture and comparison with alternative networks
• Demonstration of the open source tools used for training, optimisation and inference: TensorFlow, TensorFlow Lite, TensorFlow Lite for Microcontrollers
• Results

For whom?

The webinar is aimed at product companies and service providers active in the manufacturing industry who are developing sensors, monitoring solutions or smart systems, and having product smartification as a business driver. We will focus on R&D engineers and developers, but also welcome R&D and innovation managers.

Working language: English