arXiv 2512.166352026-03-17cs.CVcs.LG

SARMAE：用于SAR表示学习的遮蔽自动编码器

SARMAE: Masked Autoencoder for SAR Representation Learning

School of Information and Electronics, Beijing Institute of Technology, Beijing, China（北京理工大学信息与电子学院，北京，中国）；School of Computer Science, Wuhan University, Wuhan, China（武汉大学计算机学院，武汉，中国）；School of Electronic Information, Wuhan University, Wuhan, China（武汉大学电子信息学院，武汉，中国）；Zhongguancun Academy, Beijing, China（中关村学院，北京，中国）；School of Data Science, Fudan University, Shanghai, China（复旦大学数据科学学院，上海，中国）；State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, China（武汉大学测绘遥感信息工程国家重点实验室，武汉，中国）

Danxu Liu, Di Wang, Hebaixu Wang, Haoyang Chen, Wentao Jiang, Yilin Cheng, Haonan Guo, Wei Cui, Jing Zhang

备注：: The paper is accepted by CVPR 2026! Code and models will be available at https://github.com/MiliLab/SARMAE

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出SARMAE，通过构建SAR-1M数据集和引入Speckle-Aware Representation Enhancement与Semantic Anchor Representation Constraint，提升SAR图像在多任务中的表现。

摘要：

合成孔径雷达（SAR）图像在全天候、昼夜遥感应用中起关键作用。然而，现有SAR深度学习受限于数据稀缺，而SAR图像中的物理噪声进一步阻碍了细粒度语义表示学习。为此，我们提出SARMAE，一种用于自监督SAR表示学习的噪声感知遮蔽自动编码器。具体而言，我们构建了首个百万级SAR数据集SAR-1M，并配对光学图像以实现大规模预训练。在此基础上，我们设计了Speckle-Aware Representation Enhancement（SARE），将SAR特定的斑点噪声注入遮蔽自动编码器以促进噪声感知和鲁棒的表示学习。此外，我们引入了Semantic Anchor Representation Constraint（SARC），利用配对的光学先验对齐SAR特征并确保语义一致性。在多个SAR数据集上的广泛实验表明，SARMAE在分类、检测和分割任务中均取得最先进的性能。代码和模型将在https://github.com/MiliLab/SARMAE上发布。

英文摘要：

Synthetic Aperture Radar (SAR) imagery plays a critical role in all-weather, day-and-night remote sensing applications. However, existing SAR-oriented deep learning is constrained by data scarcity, while the physically grounded speckle noise in SAR imagery further hampers fine-grained semantic representation learning. To address these challenges, we propose SARMAE, a Noise-Aware Masked Autoencoder for self-supervised SAR representation learning. Specifically, we construct SAR-1M, the first million-scale SAR dataset, with additional paired optical images, to enable large-scale pre-training. Building upon this, we design Speckle-Aware Representation Enhancement (SARE), which injects SAR-specific speckle noise into masked autoencoders to facilitate noise-aware and robust representation learning. Furthermore, we introduce Semantic Anchor Representation Constraint (SARC), which leverages paired optical priors to align SAR features and ensure semantic consistency. Extensive experiments across multiple SAR datasets demonstrate that SARMAE achieves state-of-the-art performance on classification, detection, and segmentation tasks. Code and models will be available at https://github.com/MiliLab/SARMAE.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。