A collection of machine learning models for computer vision applications, audio applications and others, optimized for NXP MCUs and MPUs.
The NXP eIQ® Model Zoo offers pre-trained models for a variety of domains and tasks that are ready to be deployed on supported products.
Models are included in the form of "recipes" that convert the original models to TensorFlow Lite format. This allows users to find the original re-trainable versions of the models, allowing fine-tuning/training if required.
This should facilitate the development of embedded AI with NXP products.
Models are provided in TensorFlow Lite format.
The list of currently supported platforms can be found in the products folder.
The NXP eIQ® Model Zoo is structured in the following way: Main Page -> Domain -> Task -> Model.
Multiple models may be proposed for the same task. Each model has its own information page.
- classification
- face recognition
- object detection
- pose estimation
- semantic segmentation
- super resolution
- instance-segmentation
- low-light enhancement
The model creation recipes were tested under Ubuntu 20.04 with Python 3.8.
Copyright NXP 2024
The code in this repository is provided under MIT License.