基于排名的地理正则化：重新审视多光谱遥感图像的对比自监督学习
Rank-based Geographical Regularization: Revisiting Contrastive Self-Supervised Learning for Multispectral Remote Sensing Imagery
AI总结：
本文提出GeoRank，一种改进对比自监督学习的地理正则化方法，通过优化球面距离嵌入地理关系，提升多光谱遥感图像的特征学习效果。
摘要：
自监督学习（SSL）已成为从大规模未标记数据集中学习的强大范式，特别是在计算机视觉（CV）领域。然而，将SSL应用于多光谱遥感（RS）图像存在独特的挑战和机遇，这源于数据的地理和时间变异性。在本文中，我们介绍了GeoRank，一种新颖的对比SSL正则化方法，通过直接优化球面距离来改进先前技术，将地理关系嵌入到学习的特征空间中。GeoRank在整合地理元数据的先前方法中表现更优或持平，并且一致改进了多种对比SSL算法（例如BYOL、DINO）。此外，我们系统地研究了对比SSL在多光谱RS图像中的关键适应性，包括数据增强的有效性、数据集基数和图像大小对性能的影响，以及时间视图的任务依赖性。代码可在https://github.com/tomburgert/georank获取。
英文摘要：
Self-supervised learning (SSL) has become a powerful paradigm for learning from large, unlabeled datasets, particularly in computer vision (CV). However, applying SSL to multispectral remote sensing (RS) images presents unique challenges and opportunities due to the geographical and temporal variability of the data. In this paper, we introduce GeoRank, a novel regularization method for contrastive SSL that improves upon prior techniques by directly optimizing spherical distances to embed geographical relationships into the learned feature space. GeoRank outperforms or matches prior methods that integrate geographical metadata and consistently improves diverse contrastive SSL algorithms (e.g., BYOL, DINO). Beyond this, we present a systematic investigation of key adaptations of contrastive SSL for multispectral RS images, including the effectiveness of data augmentations, the impact of dataset cardinality and image size on performance, and the task dependency of temporal views. Code is available at https://github.com/tomburgert/georank.
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。
