Rank-based Geographical Regularization: Revisiting Contrastive Self-Supervised Learning for Multispectral Remote Sensing Imagery

2026-01-05

arXiv.org

AI Summary

This work addresses the challenge of learning effective semantic and spatial representations from multispectral remote sensing imagery, which exhibits significant spatiotemporal variability that hinders conventional self-supervised methods. To this end, we propose GeoRank, a geographically informed ranking-based regularization approach that explicitly models geographic proximity by directly optimizing spherical distances—without requiring geographic metadata. GeoRank enhances contrastive self-supervised frameworks such as BYOL and DINO by incorporating this geometric prior into their feature learning process. Comprehensive experiments across multiple remote sensing benchmarks demonstrate that GeoRank consistently improves performance across diverse self-supervised architectures, achieving results comparable to or even surpassing those of existing methods that rely on explicit geographic metadata.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Unsupervised & Self-Supervised Learning
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

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.

Problem

• self-supervised learning
• multispectral remote sensing
• geographical regularization
• contrastive learning
• geographical variability

Innovation

• GeoRank
• contrastive self-supervised learning
• geographical regularization
• multispectral remote sensing
• spherical distance