1. X-LINUX-AI v5.0.0[edit source]
Information |
This version is compatible with Yocto Project® build system Mickledore (v4.2) and has been validated against the OpenSTLinux ecosystem release v5.0.0 and validated on STM32MP1X boards. |
1.1. Contents[edit source]
- XNNPACK™ [1] as default delegate for TensorFlow Lite runtime
- TensorFlow™ Lite[2] 2.11.0
- ONNX Runtime[3] 1.14.0
- XNNPACK™[1] as Execution Provider for ONNX Runtime
- Coral Edge TPU™[4] accelerator native support
- libedgetpu 2.0.0 (Gouper) aligned with TensorFlow™ Lite 2.11.0
- libcoral 2.0.0 (Gouper) aligned with TensorFlow™ Lite 2.11.0
- PyCoral 2.0.0 (Gouper) aligned with TensorFlow™ Lite 2.11.0
- OpenCV[5] 4.7.x
- Gstreamer [6] 1.22.x
- Python™[7] 3.10.x (enabling Pillow module)
- Application samples
- C++ / Python Object detection using ONNX™ Runtime COCO SSD MobileNet v1 quantized model
- Python image classification using ONNX™ Runtime based on MobileNet v1 quantized model
- C++ / Python image classification using TensorFlow™ Lite based on MobileNet v1 quantized model
- C++ / Python object detection using TensorFlow™ Lite based on COCO SSD MobileNet v1 quantized model
- C++ / Python image classification using Coral Edge TPU™ based on MobileNet v1 quantized model and compiled for the Coral Edge TPU™
- C++ / Python object detection using Coral Edge TPU™ based on COCO SSD MobileNet v1 quantized model and compiled for the Coral Edge TPU™
- C++ face recognition using TensorFlow™ Lite models capable of recognizing the face of a known (enrolled) user (available on demand)
- Application support for all the display configurations
- C++/Python application camera pipeline update to use GtkWaylandsink and achieve better performance
- X-LINUX-AI SDK add-on extending the OpenSTLinux SDK with AI functionality to develop and build an AI application easily. The X-LINUX-AI SDK add-on provides support for all the above frameworks. It is available from the X-LINUX-AI product page
1.2. Validated hardware[edit source]
As any software expansion package, the X-LINUX-AI is supported on all STM32MP1 Series and it has been validated on the following boards: