A Multi-Agent Feedback System for Detecting and Describing News Events in Satellite Imagery
For remote sensing and journalism, this work provides a scalable method to automatically detect and describe news events in satellite imagery, addressing the lack of multi-temporal event captioning datasets.
The authors introduce SkyScraper, a multi-agent workflow that geocodes news articles and generates captions for satellite image sequences, creating a multi-temporal captioning dataset of 5,000 sequences. The system finds 5x more events than traditional geocoding methods.
Changes in satellite imagery often occur over multiple time steps. Despite the emergence of bi-temporal change captioning datasets, there is a lack of multi-temporal event captioning datasets (at least two images per sequence) in remote sensing. This gap exists because (1) searching for visible events in satellite imagery and (2) labeling multi-temporal sequences require significant time and labor. To address these challenges, we present SkyScraper, an iterative multi-agent workflow that geocodes news articles and synthesizes captions for corresponding satellite image sequences. Our experiments show that SkyScraper successfully finds 5x more events than traditional geocoding methods, demonstrating that agentic feedback is an effective strategy for surfacing new multi-temporal events in satellite imagery. We apply our framework to a large database of global news articles, curating a new multi-temporal captioning dataset with 5,000 sequences. By automatically identifying imagery related to news events, our work also supports journalism and reporting efforts.