The paper argues that satellite data could create a significant shift in machine learning, suggesting a need to rethink existing practices designed for traditional data modalities. As machine learning for satellite data (SatML) gains traction, the authors propose a new research agenda that considers the unique characteristics and challenges of satellite data. They argue that current machine learning solutions designed for other data modalities, such as natural images or language, are sub-optimal for satellite data. The paper suggests that recognizing satellite data as a distinct modality is necessary to advance the quality and impact of SatML research.
Publication date: 5 Feb 2024
Project Page: Not provided
Paper: https://arxiv.org/pdf/2402.01444
