Microelectronics

Infineon: Infineon’s DEEPCRAFT™ Studio edge AI platform now supports computer vision with Ultralytics YOLO models

February 26, 2025. Infineon Technologies AG has announced support for computer vision in DEEPCRAFT™ Studio, extending the existing offering for audio, radar and other time-series data. This enables the platform to develop edge AI models for image processing with low power consumption and storage capacities. Developers of machine learning models can use this for a wide range of applications. In an industrial environment, for example, visual inspections can be carried out in real time or machines can be switched off in the presence of a person for safety reasons. In the smart home sector, intelligent devices can monitor objects, people or pets. The use cases opened up by the use of edge AI are complex.

Share this Post
Infineon has announced support for computer vision in DEEPCRAFT™ Studio, extending the platform's existing offering with the ability to develop edge AI models for low-power and low-memory image processing. Photo: Infineon

Contact info

Silicon Saxony

Marketing, Kommunikation und Öffentlichkeitsarbeit

Manfred-von-Ardenne-Ring 20 F

Telefon: +49 351 8925 886

Fax: +49 351 8925 889

redaktion@silicon-saxony.de

Contact person:

“The new updates for DEEPCRAFT Studio optimize every step in the development of AI models for image processing-based use cases,” says Alexander Samuelsson, Product Manager DEEPCRAFT Studio. “Starting with data acquisition and data pre-processing through to model deployment. In this way, we offer developers real end-to-end support for image processing. With this support combined with Infineon’s PSOC™ Edge high-performance MCUs for machine learning, Infineon customers are well equipped to add world-class edge AI capabilities to their products.”

The new application possibilities will be presented in a computer vision demo at Embedded World 2025 in Nuremberg. It shows state-of-the-art object recognition with Ultralytics YOLO models, which make it possible to train your own computer vision solutions with DEEPCRAFT Studio. The use of Ultralytics YOLO models in DEEPCRAFT Studio significantly reduces training time while maintaining high accuracy, making the use of artificial intelligence faster and more efficient.

These enhancements to DEEPCRAFT Studio, combined with the platform’s unique graphics-based user interface, allow models to be iterated during the development cycle. Developers can run their model in real time using their own computer and camera, ensuring it works as intended before it is transferred to the hardware. This speeds up model development as improvements can be made even before it is deployed.

Availability

The new version of DEEPCRAFT Studio with support for Vision will be released in April. An automatic update will occur when the platform is restarted. Developers can download Studio here to ensure they have access to the new version as soon as it is released.

For more information on DEEPCRAFT, visit https://www.infineon.com/deepcraft

Ultralytics, a leader in artificial intelligence (AI) and developer of the Ultralytics object recognition models YOLOv5, YOLOv8 and YOLO11, continues to drive the industry forward with more than 72m Pip downloads, more than 100,000 GitHub stars and more than 5,000 research citations, is driving the development of state-of-the-art computer vision solutions across industries and making AI faster, smarter and more accessible worldwide.

About Infineon

Infineon Technologies AG is a leading global provider of semiconductor solutions for power systems and the Internet of Things (IoT). With its products and solutions, Infineon is driving decarbonization and digitalization. The company has around 58,060 employees worldwide (end of September 2024) and generated sales of around 15 billion euros in the 2024 financial year (end of September).

– – – – – –

Further links

👉 www.infineon.com  

Photo: Infineon

You may be interested in the following

Contact info

Silicon Saxony

Marketing, Kommunikation und Öffentlichkeitsarbeit

Manfred-von-Ardenne-Ring 20 F

Telefon: +49 351 8925 886

Fax: +49 351 8925 889

redaktion@silicon-saxony.de

Contact person: