In crisis situations, emergency responders and aid organizations require timely and accurate information on hazard extent, exposed assets, and resulting damage to coordinate effective response operations. Remote sensing has long played a crucial role in disaster management: the increasing availability of satellite, airborne, and UAV data, combined with enhanced spatial, spectral, and temporal resolution, now allows for near-real-time and global-scale monitoring of vulnerable regions. However, turning this wealth of data into actionable information remains a major challenge. Recent advances in artificial intelligence (AI) offer promising solutions by automating the extraction of critical information from remotely sensed data, paving the way to real-time, large-scale impact assessments.

This session invites contributions on innovative methods for automatic and scalable information retrieval from remote sensing data in crisis contexts, including: (i) AI models for damage detection and impact assessment, (ii) multi-sensor, multi-platform data integration for scalable disaster analysis, (iii) UAV-based data for inaccessible or rapidly changing disaster zones, (iv) real-time processing, alert systems, and rapid mapping and, (v) data standards, benchmark datasets, and best practices for algorithm training and evaluation.