The Edge AI development platform, DEEPCRAFT Studio, is expanding to include computer vision capabilities alongside its existing audio, radar, and time-series data support. This addition enables developers to create Edge AI models for vision applications with minimal power and memory requirements.
The updated platform provides comprehensive support for vision-based AI model development, covering the entire process from data collection and preprocessing to model deployment. When paired with high-performance machine learning microcontrollers, developers can implement Edge AI features in their products.
This technology will be demonstrated at Embedded World 2025 in Nuremberg this March. The demonstration will feature object detection powered by Ultralytics YOLO models, which allow developers to train custom computer vision solutions. These models reduce training time while maintaining the accuracy levels necessary for efficient AI deployment.
The platform includes a graph-based user interface that facilitates model iteration during development. Developers can test models in real-time using their computer and camera before hardware deployment, allowing for adjustments and improvements before final implementation.
The updated version of DEEPCRAFT Studio with vision support will be available in April. The platform updates automatically upon restart so that current users will receive the new functionality without additional steps.
Leave a Reply