INSTRUMENTATION AND ELECTRONICS
CLEAR VISION A
Infi neon has expanded the capabilities of its AI Deepcraft studio to include computer vision
I has many diff erent applications, and one of them is sight. This is a potential gamechanger
for machine learning developers as it means a direct hit of information, less coding and a more ‘human’ means of interacting with the environment. With this in mind, business solutions
company Infi neon Technologies has released support for computer vision in its Deepcraft Studio - a platform that enables end-to-end development of Edge AI applications for embedded devices. This addition expands on the existing support for audio, radar and other time-series data. The addition is being supported by
Ultralytics, an AI leader and creator of YOLOv5, YOLOv8, and YOLO11 object detection models. The product powers ‘cutting-edge computer vision across industries with 72M+ Pip downloads, 100K+ GitHub stars, and 5,000+ research citations,’ according to the company. With the addition, the platform can
now develop low-power, low-memory Edge AI models for vision. Use cases for this updated AI implementation will include factories wanting to conduct real-time visual inspection of parts, for example, or shut down machinery in the presence of a human; or Smart Home appliances looking to monitor objects, humans, or pets; for example. Alexander Samuelsson, product
manager Deepcraft Studio said: “These updates will optimise each step in AI model development for vision-based use cases, starting with data collection and preprocessing and all the way through model deployment,” he continued, “this means we are off ering developers end-to- end support for vision. When coupled with Infi neon’s PSOC Edge high performance machine learning MCUs, our customers will be well-equipped to add a range of high end Edge AI features to their products.”
50
www.engineerlive.com
Infi nion’s Deepcraft Studio now supports vision
Infi neon will showcase this
capability in a computer vision demo at Embedded World 2025 in Nuremberg this March. The detection will be powered by Ultralytics YOLO models, enabling developers to train custom computer vision solutions. Leveraging Ultralytics Yolo models will signifi cantly reduce training time while maintaining high accuracy. These additions to the Deepcraft
Studio paired with the platform’s unique graph-based UX interface enable iterations on models during the development cycle. Developers can run their model in real-time using their
own computer and camera, making sure it works as intended before deploying onto hardware. This speeds up development, as improvements can be made to the model before it is deployed.
AVAILABILITY Infi neon will launch a new version of the studio with support for vision in April.
For more information visit:
www.infi
neon.com/deepcraft
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56
