PulseAugur
LIVE 04:15:31
tool · [1 source] ·
0
tool

GeoQuery system uses text proxies for zero-shot satellite image retrieval in crisis response

Researchers have developed GeoQuery, a novel zero-shot retrieval system designed for searching vast archives of satellite imagery, particularly for crisis response applications. This system bypasses the need for extensive paired data and compute by using prompt-aligned text proxies to bridge the gap between visual foundation models like CLAY and natural language queries. GeoQuery demonstrated strong performance on disaster-related queries, achieving significant accuracy in identifying vulnerable areas during simulated and real-world events, and has been integrated into the ECHO crisis response system. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Offers a practical method for operationalizing large-scale Earth observation data retrieval for crisis response when full contrastive training is infeasible.

RANK_REASON Academic paper introducing a new system for satellite image retrieval. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

COVERAGE [1]

  1. arXiv cs.CV TIER_1 · James Walsh, William Fawcett, Grace Colvard, Ra\'ul Ramos-Poll\'an ·

    Zero-Shot Satellite Image Retrieval through Joint Embeddings: Application to Crisis Response

    arXiv:2605.05405v1 Announce Type: new Abstract: Semantic search of Earth observation archives remains challenging. Visual foundation models such as CLAY produce rich embeddings of satellite imagery but lack the natural-language grounding needed for intuitive query, and full contr…