X-LINUX-AI v6.1.0

Signed-off-by: Maxence GUILHIN <[email protected]>

Author: mguSTM

LICENSE.md

@@ -13,13 +13,16 @@
 | python3-libstai-mpu                                                          | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | python3-libtensorflow-lite                                                   | The TensorFlow Authors                                                | [Apache-2.0](https://opensource.org/license/apache-2-0)
 | pyton3-onnxruntime                                                           | Advanced Micro Devices, Inc, Alex Brick, Amazon.com, Inc. or its affiliates, Apple Inc, ARM Ltd, Baptiste Wicht, Broadcom Limited, Bull SAS, Chelsio, Inc, Cisco Systems, Inc, DataDirect Networks, Deepmind Technologies (Koray Kavukcuoglu), Dmitry Vyukov, Emscripten authors (see AUTHORS in Emscripten), Evergrid, Inc, Facebook, Inc. and its affiliates, Forschungszentrum Juelich, Juelich Supercomputing, Fujitsu Limited, Georgia Institute of Technology, Google LLC, HalideIR contributors, Hewlett-Packard Development Company, LP, High Performance Computing Center Stuttgart, Huawei Technologies Co., Ltd, IBM Corporation, Idiap Research Institute (Ronan Collobert, Samy Bengio, Johnny Mariethoz), INADA Naoki, Institut National de Recherche en Informatique, Intel Corporation, Jouni Malinen and contributors, Junio C Hamano, Lawrence Livermore National Security, LLC, Lisandro Dalcin., Los Alamos National Security, LLC, Marat Dukhan, Martin Moene, Matthew Stevens, Mellanox Technologies Ltd, Microsoft, Microsoft Corporation, MIT CSAIL, Google Inc., and other contributors, MITSUNARI Shigeo, Myricom, Inc., NEC Laboratories America (Koray Kavukcuoglu ,Ronan Collobert, Leon Bottou, Iain Melvin, Jason Weston), Nicola Iarocci, Niels Lohmann, NumPy Developers, NVIDIA CORPORATION & AFFILIATES, NXP Semiconductor, Inc, NYU (Clement Farabet), Oak Ridge National Labs, Oliver Nightingale, Open Neural Network Exchange, Oracle and/or its affiliates, QLogic Corporation, Research Organization for Information Science (RIST), Sandia National Laboratories, Stony Brook University, Sun Microsystems, Inc, The Android Open Source Project, The RE2 Authors, The Regents of the University of California (Regents), the respective contributors, The scikit-learn developers, The Trustees of Indiana University and Indiana, The University of Houston System, The University of Tennessee and The University, The University of Wisconsin-La Crosse, Triad National Security, LLCMyricom, Tzu-Wei Huang, UChicago Argonne, LLC, UT-Battelle, LLC, Voltaire, Inc, Wenzel Jakob, Xilinx Inc, Yangqing Jia, ZIH, TU Dresden, Federal Republic of Germany | [MIT](https://opensource.org/license/mit)
+| python3-onnx                                                                 | ONNX Project Contributors, Sphinx Project, Rodrigo Fernandes, Junio C Hamano | [Apache-2.0](https://opensource.org/license/apache-2-0)
+| python3-torch                                                                | Pytorch Team, Rich Geldreich and Tenacious Software LLC, Stephan Brumme, The Android Open Source Project, Meta Platforms, Inc. and its affiliates, Facebook, Inc., Microsoft Corporation, Google Inc., Martin Raiber, Advanced Micro Devices, Inc., Mark Dickinson, Robert Kern, RAD Game Tools and Valve Software, Jeff Preshing, Julian Seward, NVIDIA CORPORATION & AFFILIATES, Cameron Desrochers, Python Software Foundation, Tri Dao, NumPy Developers, SymPy Development Team, Cruise LLC, Contributors, Facebook Inc., Giovanni Garberoglio, Eddy Petrișor, The TensorFlow Authors, Kakao Brain, MathInf GmbH, Thomas Viehmann, Edward Z. Yang, manylinux, Indiana University, Massachusetts Institute of Technology, The ggml authors, ActiveState Software Inc., Steven Moshier, Alex Rogozhnikov, Melissa E. O'Neill, Björn Fahller, Norbert Juffa, Marat Dukhan, Josef Weidendorfer, Makoto Matsumoto and Takuji Nishimura, Georgia Institute of Technology, Donald Stufft and individual contributors | [BSD-3-Clause](https://opensource.org/licenses/BSD-3-Clause)
 | tim-vx                                                                       | VeriSilicon Microelectronics(Shanghai) Co., Ltd.                      | [MIT](https://opensource.org/license/mit)
 | tflite-vx-delegate                                                           | VeriSilicon Microelectronics(Shanghai) Co., Ltd.                      | [MIT](https://opensource.org/license/mit)
 
 ## Tools
 | Components                                                                   | Copyright                                                             | License
 |:---------                                                                    |:----------                                                            |:-------
-| apt-openstlinux-x-linux-ai                                                   | STMicroelectronics                                                    | [MIT](https://opensource.org/license/mit)
+| apt-openstlinux-x-linux-ai-cpu                                               | STMicroelectronics                                                    | [MIT](https://opensource.org/license/mit)
+| apt-openstlinux-x-linux-ai-npu                                               | STMicroelectronics                                                    | [MIT](https://opensource.org/license/mit)
 | x-linux-ai-benchmark                                                         | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | x-linux-ai-tool                                                              | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | x-linux-ai-application                                                       | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
@@ -36,6 +39,8 @@
 | application-resources                                                        | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | config-cpu                                                                   | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | config-npu                                                                   | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
+| odl-teacher-student-object-detection-jupyterlab                              | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
+| odl-teacher-student-object-detection-python                                  | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | stai-mpu-image-classification-cpp                                            | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | stai-mpu-image-classification-cpp-ort-cpu                                    | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
 | stai-mpu-image-classification-cpp-ort-npu                                    | STMicroelectronics                                                    | [SLA0044](https://www.st.com/SLA0044)
@@ -76,6 +81,8 @@
 | models image classification: mobilenetv2-10-224                              | STMicroelectronics                                                    | [Apache-2.0](https://opensource.org/license/apache-2-0)
 | models object detection: ssd-mobilenetv1-10-300                              | Google                                                                | [Apache-2.0](https://opensource.org/license/apache-2-0)
 | models object detection: ssd-mobilenetv2-10-256-fpnlite                      | STMicroelectronics                                                    | [Apache-2.0](https://opensource.org/license/apache-2-0)
+| models object detection On device Learning: ssd-mobilenetv2-10-300           | STMicroelectronics                                                    | [Apache-2.0](https://opensource.org/license/apache-2-0)
+| models object detection On device Learning: rtdetr-large-256                 | Ultralytics, STMicroelectronics                                       | [AGPL-3.0-only](https://opensource.org/license/agpl-v3)*
 | models people detection: Yolov8n                                             | Ultralytics, STMicroelectronics                                       | [AGPL-3.0-only](https://opensource.org/license/agpl-v3)*
 | models pose estimation: Yolov8n                                              | Ultralytics, STMicroelectronics                                       | [AGPL-3.0-only](https://opensource.org/license/agpl-v3)*
 | models semantic segmentation: deeplabv3-257                                  | Google                                                                | [Apache-2.0](https://opensource.org/license/apache-2-0)

README.md

@@ -2,7 +2,7 @@
     <img width="720" src="https://raw.githubusercontent.com/STMicroelectronics/meta-st-stm32mpu-ai/master/x-linux-ai-logo.png">
 </p>
 
-X-LINUX-AI version: v6.0.1
+X-LINUX-AI version: v6.1.0
 
 X-LINUX-AI is a free of charge open-source software package dedicated to AI.
 It is a complete ecosystem that allow developers working with OpenSTLinux to create AI-based application very easily.
@@ -19,8 +19,8 @@ X-LINUX-AI OpenEmbedded meta layer to be integrated into OpenSTLinux distributio
 It contains recipes for AI frameworks, tools and application examples for STM32MPx series
 
 ## Compatibility
-The X-LINUX-AI OpenSTLinux Expansion Package v6.0.1 is compatible with the Yocto Project™ build system Scarthgap.
-It is validated over the OpenSTLinux Distribution v6.0.0 https://wiki.st.com/stm32mpu on STM32MP25x and STM32MP1x series.
+The X-LINUX-AI OpenSTLinux Expansion Package v6.1.0 is compatible with the Yocto Project™ build system Scarthgap.
+It is validated over the OpenSTLinux Distribution v6.1.0 https://wiki.st.com/stm32mpu on STM32MP2x and STM32MP1x series.
 
 ## Versioning
 Since its release v5.0.0, the major versioning of the X-LINUX-AI OpenSTLinux Expansion Package is aligned on the major versioning of the OpenSTLinux Distribution. This prevents painful backward compatibility attempts and makes dependencies straightforward.
@@ -30,12 +30,14 @@ The X-LINUX-AI generic versioning v**x**.**y**.**z** is built as follows:
 * **z**: patch version to introduce bug fixes. A patch version is implemented in a backward compatible manner.
 
 ## Available frameworks and tools within the meta-layer
-[X-LINUX-AI v6.0.1 expansion package](https://wiki.st.com/stm32mpu/wiki/Category:X-LINUX-AI_expansion_package):
+[X-LINUX-AI v6.1.0 expansion package](https://wiki.st.com/stm32mpu/wiki/Category:X-LINUX-AI_expansion_package):
 * AI Frameworks:
-  * STAI_MPU Unified API based on OpenVX™(STM32MP25x only), TensorFlow™ Lite, and ONNX Runtime™ compatible with all STM32MPU series
-  * TIM-VX™ 1.2.6 (STM32MP25x only)
-  * TensorFlow™ Lite 2.16.2 with XNNPACK delegate activated for CPU execution and VX-delegate External delegate to address STM32MP2 NPU
-  * ONNX Runtime™ 1.19.2 with XNNPACK execution engine activated for CPU execution and VSINPU Execution provider to address STM32MP2 NPU
+  * STAI_MPU Unified API based on OpenVX™(STM32MP25x and STM32MP23x only), TensorFlow™ Lite, and ONNX Runtime™ compatible with all STM32MPU series
+  * TIM-VX™ 1.22.6 (STM32MP25x and STM32MP23x only)
+  * TensorFlow™ Lite 2.18.0 with XNNPACK delegate activated for CPU execution and VX-delegate External delegate to address STM32MP2 NPU
+  * ONNX Runtime 1.19.2 with XNNPACK execution engine activated for CPU execution and VSINPU Execution provider to address STM32MP2 NPU
+  * ONNX 1.16.2 python version for On Device Learning
+  * Pytorch 2.3.1 python version for On Device Learning
 
 * Out of the box applications:
   * Image classification :
@@ -48,6 +50,11 @@ The X-LINUX-AI generic versioning v**x**.**y**.**z** is built as follows:
     * Python™ example using STAI_MPU Unified API based on DeepLabV3 quantized model
   * Face recognition:
     * C++ example using STAI_MPU unified API based on the BlazeFace and FaceNet quantized models
+  * People Tracking and Heatmap
+    * Python™ example using STAI_MPU Unified API based on the yolov8n quantized model.
+  * On Device Learning for Object detection
+    * Python™ example using STAI_MPU Unified API based on SSD MobileNet v2 as student and RT-DETR transformer model as teacher.
+    * Step by step Jupyterlab™ notebook available for this application.
   * Note: applications are based on Gstreamer 1.22.x, GTK 3.x, OpenCV 4.9.x, Pillow, Python 3
 
 * Utilities:
@@ -68,5 +75,8 @@ The X-LINUX-AI generic versioning v**x**.**y**.**z** is built as follows:
 ## Further information on how to install and how to use X-LINUX-AI Distribution package
 <https://wiki.st.com/stm32mpu/wiki/X-LINUX-AI_Distribution_package>
 
+## Further information on On Device Learning feature
+<https://wiki.st.com/stm32mpu/wiki/Category:AI_-_On_Device_Learning>
+
 ## Application samples
 <https://wiki.st.com/stm32mpu/wiki/Category:AI_-_Application_examples>