Prominent Notice

Release of Source Code and Trained Models: License Compliance.

To ensure compliance with the AGPL 3.0 license and Apache 2.0 all source code and trained models for this project have been released and are available for download. This includes all modifications made compared to the base version of the code. 

Please note that some parts of the code are under specific licenses. We encourage users to refer to those licenses for details on usage and distribution.

This project includes components licensed under:

- Apache License 2.0 (for Axis components)

- AGPL 3.0 (for Ultralytics YOLOv5 components)

As the project incorporates AGPL 3.0 licensed components the source code should be available to users if you distribute the software, and any modifications must also be shared under the same license. Additionally, since the project uses components licensed under the Apache License 2.0, we encourage you to comply with the terms of that license for those components.

A copy of the AGPL 3.0 license and Apache 2.0 license are included in the root folder for reference.

This project is made up of 3 components:

- 0_model_section: yolov5 based AI model trained using a Bottom Up Domain Layer Adaptation on thermal images

- 1_tracing_section: how the trained pytorch model is converted for executing in C++ edge thermal camera

- 2_edge_application: the code of the application specific for executing in AXIS camera environment

(For compliance regarding the Axis components, please refer to the applicable licenses)

For each section it is associated a folder with the code or README containing the instructions to reproduce the project

The case study of this project used images from on images from Palazzo Vecchio, Florence Italy. More details can be found @ https://link.springer.com/article/10.1007/s11042-023-16374 (Last Accessed Date: 01/10/2024). But you can choose and annotate the images for your custom application scenario. 

We recommend reviewing the updated documentation to fully understand the changes and how to integrate them into your workflow.

Community contributions are always welcome. If you have any questions or need further assistance reach out to us via mail at: info@d-ai.eu

Thank you for your support!

Last update: 01/10/2024