C-DiffSET: Leveraging Latent Diffusion for SAR-to-EO Image Translation With Confidence-Guided Reliable Object Generation
IEEE Transactions on Circuits and Systems for Video Technology · 2026-10 · 36 (10) · 14604-14620
DOI: 10.1109/tcsvt.2026.3701447 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-10. Verfasst von Jeonghyeok Do und 3 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 105 Literaturangaben. Gefördert von: National Research Foundation of Korea, Korean Government [Ministry of Science and Information and Communications Technology (ICT)].
Abstract (Originalsprache)
Synthetic Aperture Radar (SAR) imagery provides robust environmental and temporal coverage (e.g., during clouds, seasons, day-night cycles), yet its noise and unique structural patterns pose interpretation challenges, especially for non-experts. SAR-to-EO (Electro-Optical) image translation (SET) has emerged to make SAR images more perceptually interpretable. However, traditional approaches, trained from scratch on limited SAR-EO datasets, are prone to overfitting and struggle with dataset inconsistencies. To address these challenges, we introduce Confidence Diffusion for SAR-to-EO Translation (C-DiffSET), a framework leveraging a pretrained Latent Diffusion Model (LDM) to effectively adapt its extensive generative priors from natural images to the EO domain. Our inves-tigation reveals that the pretrained VAE encoder effectively aligns SAR and EO images within a shared latent space, demonstrating robustness even to varying noise levels in SAR inputs. To further improve pixel-wise fidelity for SET and mitigate artifacts from temporal discrepancies, such as appearing or disappearing objects, we propose a novel confidence-guided diffusion (C-Diff) loss. This loss dynamically guides the diffusion process to down-weight penalties in uncertain regions, thereby enhancing structural accuracy. C-DiffSET achieves state-of-the-art (SOTA) results on multiple benchmark datasets (QXS-SAROPT, SAR2Opt, SpaceNet6, Stellar-Vision), significantly outperforming recent image-to-image translation methods and specialized SET methods across all standard metrics. The source code and trained models are available at https://github.com/KAIST-VICLab/C-DiffSET
Förderung
- National Research Foundation of Korea
- Korean Government [Ministry of Science and Information and Communications Technology (ICT)]
Zitieren
APA
Do, J., Lee, J., Lee, S., & Kim, M. (2026). C-DiffSET: Leveraging Latent Diffusion for SAR-to-EO Image Translation With Confidence-Guided Reliable Object Generation. IEEE Transactions on Circuits and Systems for Video Technology, 36(10), 14604-14620. https://doi.org/10.1109/tcsvt.2026.3701447
Vancouver
Do J, Lee J, Lee S, Kim M. C-DiffSET: Leveraging Latent Diffusion for SAR-to-EO Image Translation With Confidence-Guided Reliable Object Generation. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(10):14604-14620. doi:10.1109/tcsvt.2026.3701447
BibTeX
@article{Do2026,
author = {Do, Jeonghyeok and Lee, Jaehyup and Lee, Seungchul and Kim, Munchurl},
title = {C-DiffSET: Leveraging Latent Diffusion for SAR-to-EO Image Translation With Confidence-Guided Reliable Object Generation},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2026},
volume = {36},
number = {10},
pages = {14604--14620},
doi = {10.1109/tcsvt.2026.3701447}
}
EUVIMED