RSAR：受限状态角度解析器与旋转SAR基准
RSAR: Restricted State Angle Resolver and Rotated SAR Benchmark
AI总结：
针对SAR旋转目标检测缺乏大规模数据集及弱监督角度预测不准的问题，提出引入单位圆约束的UCR，并据此构建最大多类别旋转SAR数据集RSAR。
摘要：
旋转目标检测在光学遥感领域已取得显著进展。然而，合成孔径雷达（SAR）领域的进展相对滞后，主要原因在于缺乏大规模数据集。标注此类数据集效率低且成本高。一种有前景的解决方案是采用弱监督模型（例如，仅利用可用的水平框进行训练）生成伪旋转框，以供人工校准前参考。遗憾的是，现有弱监督模型在预测目标角度方面准确性有限。以往工作尝试通过角度解析器将角度解耦为余弦和正弦编码来增强角度预测。本文首先从维度映射的统一视角重新评估这些解析器，并揭示它们具有相同的缺点：这些方法忽视了这些编码中固有的单位周期约束，容易导致预测偏差。为解决该问题，我们提出单位周期解析器（Unit Cycle Resolver，UCR），其引入单位圆约束损失以提高角度预测准确性。我们的方法能够有效提升现有最先进弱监督方法的性能，甚至在现有光学基准（即DOTA-v1.0数据集）上超越全监督模型。借助UCR，我们进一步标注并推出RSAR，这是迄今为止最大的多类别旋转SAR目标检测数据集。在RSAR和光学数据集上的大量实验表明，我们的UCR提升了角度预测准确性。数据集和代码可在https://github.com/zhasion/RSAR获取。
英文摘要：
Rotated object detection has made significant progress in the optical remote sensing. However, advancements in the Synthetic Aperture Radar (SAR) field are laggard behind, primarily due to the absence of a large-scale dataset. Annotating such a dataset is inefficient and costly. A promising solution is to employ a weakly supervised model (e.g., trained with available horizontal boxes only) to generate pseudo-rotated boxes for reference before manual calibration. Unfortunately, the existing weakly supervised models exhibit limited accuracy in predicting the object's angle. Previous works attempt to enhance angle prediction by using angle resolvers that decouple angles into cosine and sine encodings. In this work, we first reevaluate these resolvers from a unified perspective of dimension mapping and expose that they share the same shortcomings: these methods overlook the unit cycle constraint inherent in these encodings, easily leading to prediction biases. To address this issue, we propose the Unit Cycle Resolver, which incorporates a unit circle constraint loss to improve angle prediction accuracy. Our approach can effectively improve the performance of existing state-of-the-art weakly supervised methods and even surpasses fully supervised models on existing optical benchmarks (i.e., DOTA-v1.0 dataset). With the aid of UCR, we further annotate and introduce RSAR, the largest multi-class rotated SAR object detection dataset to date. Extensive experiments on both RSAR and optical datasets demonstrate that our UCR enhances angle prediction accuracy. Our dataset and code can be found at: https://github.com/zhasion/RSAR.
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。
