GeoDiff-SAR: A Geometric Prior Guided Diffusion Model for SAR Image Generation

TL;DR

GeoDiff-SAR introduces a physics-guided diffusion model that incorporates geometric priors and multi-modal fusion to generate high-fidelity SAR images with precise control over parameters, improving downstream classification accuracy.

Contribution

The paper presents a novel SAR image generation method combining geometric priors, feature fusion, and lightweight fine-tuning of a diffusion model, addressing limitations of existing image-only approaches.

Findings

Generated images exhibit high fidelity and realism.

Significant improvement in classification accuracy across azimuth angles.

Effective integration of physical SAR models into generative processes.

Abstract

Synthetic Aperture Radar (SAR) imaging results are highly sensitive to observation geometries and the geometric parameters of targets. However, existing generative methods primarily operate within the image domain, neglecting explicit geometric information. This limitation often leads to unsatisfactory generation quality and the inability to precisely control critical parameters such as azimuth angles. To address these challenges, we propose GeoDiff-SAR, a geometric prior guided diffusion model for high-fidelity SAR image generation. Specifically, GeoDiff-SAR first efficiently simulates the geometric structures and scattering relationships inherent in real SAR imaging by calculating SAR point clouds at specific azimuths, which serves as a robust physical guidance. Secondly, to effectively fuse multi-modal information, we employ a feature fusion gating network based on Feature-wise Linear Modulation (FiLM) to dynamically regulate the weight distribution of 3D physical information, image control parameters, and textual description parameters. Thirdly, we utilize the Low-Rank Adaptation (LoRA) architecture to perform lightweight fine-tuning on the advanced Stable Diffusion 3.5 (SD3.5) model, enabling it to rapidly adapt to the distribution characteristics of the SAR domain. To validate the effectiveness of GeoDiff-SAR, extensive comparative experiments were conducted on real-world SAR datasets. The results demonstrate that data generated by GeoDiff-SAR exhibits high fidelity and effectively enhances the accuracy of downstream classification tasks. In particular, it significantly improves recognition performance across different azimuth angles, thereby underscoring the superiority of physics-guided generation.