SARFormer -- An Acquisition Parameter Aware Vision Transformer for Synthetic Aperture Radar Data
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866915236945068032 |
|---|---|
| author | Prexl, Jonathan Recla, Michael Schmitt, Michael |
| author_facet | Prexl, Jonathan Recla, Michael Schmitt, Michael |
| contents | This manuscript introduces SARFormer, a modified Vision Transformer (ViT) architecture designed for processing one or multiple synthetic aperture radar (SAR) images. Given the complex image geometry of SAR data, we propose an acquisition parameter encoding module that significantly guides the learning process, especially in the case of multiple images, leading to improved performance on downstream tasks. We further explore self-supervised pre-training, conduct experiments with limited labeled data, and benchmark our contribution and adaptations thoroughly in ablation experiments against a baseline, where the model is tested on tasks such as height reconstruction and segmentation. Our approach achieves up to 17% improvement in terms of RMSE over baseline models |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_08441 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SARFormer -- An Acquisition Parameter Aware Vision Transformer for Synthetic Aperture Radar Data Prexl, Jonathan Recla, Michael Schmitt, Michael Computer Vision and Pattern Recognition This manuscript introduces SARFormer, a modified Vision Transformer (ViT) architecture designed for processing one or multiple synthetic aperture radar (SAR) images. Given the complex image geometry of SAR data, we propose an acquisition parameter encoding module that significantly guides the learning process, especially in the case of multiple images, leading to improved performance on downstream tasks. We further explore self-supervised pre-training, conduct experiments with limited labeled data, and benchmark our contribution and adaptations thoroughly in ablation experiments against a baseline, where the model is tested on tasks such as height reconstruction and segmentation. Our approach achieves up to 17% improvement in terms of RMSE over baseline models |
| title | SARFormer -- An Acquisition Parameter Aware Vision Transformer for Synthetic Aperture Radar Data |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.08441 |