Compilation and Execution of an Embeddable YOLO-NAS on the VTA
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918470015254528 |
|---|---|
| author | Faure-Gignoux, Anthony Delmas, Kevin Gauffriau, Adrien Pagetti, Claire |
| author_facet | Faure-Gignoux, Anthony Delmas, Kevin Gauffriau, Adrien Pagetti, Claire |
| contents | Deploying complex Convolutional Neural Networks (CNNs) on FPGA-based accelerators is a promising way forward for safety-critical domains such as aeronautics. In a previous work, we have explored the Versatile Tensor Accelerator (VTA) and showed its suitability for avionic applications. For that, we developed an initial stand-alone compiler designed with certification in mind. However, this compiler still suffers from some limitations that are overcome in this paper. The contributions consist in extending and fully automating the VTA compilation chain to allow complete CNN compilation and support larger CNNs (which parameters do not fit in the on-chip memory). The effectiveness is demonstrated by the successful compilation and simulated execution of a YOLO-NAS object detection model. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_24455 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Compilation and Execution of an Embeddable YOLO-NAS on the VTA Faure-Gignoux, Anthony Delmas, Kevin Gauffriau, Adrien Pagetti, Claire Hardware Architecture Deploying complex Convolutional Neural Networks (CNNs) on FPGA-based accelerators is a promising way forward for safety-critical domains such as aeronautics. In a previous work, we have explored the Versatile Tensor Accelerator (VTA) and showed its suitability for avionic applications. For that, we developed an initial stand-alone compiler designed with certification in mind. However, this compiler still suffers from some limitations that are overcome in this paper. The contributions consist in extending and fully automating the VTA compilation chain to allow complete CNN compilation and support larger CNNs (which parameters do not fit in the on-chip memory). The effectiveness is demonstrated by the successful compilation and simulated execution of a YOLO-NAS object detection model. |
| title | Compilation and Execution of an Embeddable YOLO-NAS on the VTA |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2604.24455 |