EO2SAR-Diff: Structure-Aware Latent Diffusion for Unpaired EO-to-SAR Translation.
Saved in:
| Title: | EO2SAR-Diff: Structure-Aware Latent Diffusion for Unpaired EO-to-SAR Translation. |
|---|---|
| Authors: | Kim, Yeon-Wook1 (AUTHOR) yw2kim@infiniq.co.kr, Kim, Kiyoung1 (AUTHOR) |
| Source: | Remote Sensing. Jun2026, Vol. 18 Issue 12, p2037. 35p. |
| Subjects: | Synthetic aperture radar, Data augmentation, Object recognition (Computer vision), Remote sensing, Image processing |
| Abstract: | Highlights: What are the main findings? EO2SAR-Diff, a conditional latent diffusion framework, ranks in the top tier in distributional alignment with real SAR imagery, in FID and KID computed with SAR-domain-adapted feature extractors, among GAN- and diffusion-based comparison methods. The framework preserves the spatial structure of EO inputs while reproducing SAR-specific scattering characteristics such as speckle noise, texture patterns, and rim scattering signatures. What are the implications of the main findings? Decoupling structural preservation from SAR-specific style transfer and modulating it along both spatial and temporal axes constitutes an effective design principle for cross-modal image translation between heterogeneous modalities such as EO and SAR. Augmenting training data with EO2SAR-Diff outputs yields consistent improvements in downstream detection performance, demonstrating the framework's utility as a scalable data augmentation tool for mitigating SAR data scarcity. Synthetic aperture radar (SAR) imagery provides all-weather, day-and-night observation capabilities that complement electro-optical (EO) imaging; however, the limited number of operational SAR satellites and the difficulty of acquiring expert-annotated SAR datasets constrain deep-learning-based SAR image analysis. In this paper, we propose EO2SAR-Diff, a conditional latent diffusion framework that translates EO aerial images into realistic synthetic SAR images. The framework comprises three core components: (1) domain-adaptive LoRA pre-training that anchors the Stable Diffusion backbone in the remote sensing domain, (2) a style extraction and injection network that captures SAR-specific visual characteristics via multi-scale feature encoding and parallel cross-attention, and (3) a multi-branch ControlNet with three parallel branches for complementary structural guidance. These components are coordinated by a dual-axis feature injection strategy that modulates conditioning strength along both spatial (per-block) and temporal (per-timestep) dimensions. Experiments on the DOTA 1.0 and SARDet-100K datasets demonstrate that EO2SAR-Diff ranks in the top tier among all compared methods in distributional alignment with real SAR imagery, in terms of FID and KID computed with two SAR-domain-adapted feature extractors. Augmenting the SAR training set with our synthetic images yields consistent improvements in downstream object detection performance, confirming the practical utility of the proposed framework. [ABSTRACT FROM AUTHOR] |
| Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
Be the first to leave a comment!
