发表机构
British Antarctic Survey（英国南极调查局）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
该研究提出基于互信息对齐的架构，对齐Sentinel-1与MODIS的多模态卫星场景，构建稀疏专家标注数据集，提升海冰分类准确率，实现从稀疏监督的精确密集海冰分割。
AI 中文摘要
多模态卫星影像为地球观测提供互补信息，但在动态环境中准确融合异构传感器数据仍具挑战性。南极边缘冰区等快速变化区域无法充分利用不同卫星传感器的多模态信息，因为地表特征在图像采集期间会发生移动。这种时空失配会阻碍有效的感知定位，违背了大多数多模态推理及下游分类流程所依赖的像素级对应假设。南极海冰是极具挑战性的基准测试对象，因为单个浮冰会快速且不均匀地漂移，且海冰对雷达、可见光和热传感模态的响应存在差异。海冰的精确密集标注数据仍然稀缺，生成像素级标签需要专家对噪声数据进行耗时的解读，导致模型训练长期依赖低分辨率的海事冰图。本文提出一种基于互信息对齐的新型架构，用于对齐多卫星平台（Sentinel-1与MODIS）的多模态卫星场景（涵盖可见光、热、雷达模态）。为验证该方法，我们引入了包含2088个像素级标注（7046个专家点分类）的稀疏专家标注数据集，这些标注分布在43个场景的冰-水边缘界面。实验结果表明，在分割前对模态进行空间定位与对齐可提升分类准确率，并能从稀疏点级监督中实现精确密集的海冰分割。
英文摘要
Multimodal satellite imagery provides complementary information for Earth Observation, but accurately combining heterogeneous sensors remains challenging in dynamic environments. Fast-changing regions, such as the Antarctic marginal ice zone, cannot fully exploit multimodal information from different satellite sensors because surface features move between image acquisitions. This spatial and temporal mismatch challenges effective perceptual grounding, violating the assumption of pixel-level correspondence that underpins most multimodal reasoning and downstream classification pipelines. Antarctic sea ice provides a challenging benchmark due to the rapid, heterogeneous drift of individual ice floes and the differing responses of sea ice to radar, visible and thermal sensing modalities. Accurate, dense supervision of sea ice remains scarce because generating pixel-wise labels requires time-consuming expert interpretation of noisy data, leading to historical reliance on coarse-resolution maritime ice charts for model training. This paper presents a novel architecture based on mutual information warping to align multi-satellite (Sentinel-1 and MODIS platforms) multimodal (visible, thermal, radar) satellite scenes. To demonstrate the approach, we introduce a sparse expert-labeled dataset of 2,088 pixel-wise annotations (7,046 expert point classifications) located at the ice-water margin interface across 43 scenes. Our results demonstrate that spatially grounding and aligning modalities prior to segmentation improves classification accuracy, and enables accurate, dense sea ice segmentation from sparse point-wise supervision.
CommentsECCV Workshop paper; BEAM 2
