Foundation Model for Efficient Natural Disaster Damage Assessment Using Multi-Modal Data

Watch a 20-minute conference presentation exploring innovative approaches to natural disaster damage assessment using artificial intelligence and multimodal data analysis. Learn how Associate Professor Maryam Rahnemoonfar from Lehigh University addresses the increasing challenges posed by climate change-induced natural disasters through advanced technological solutions. Discover the advantages of using Small Unmanned Aerial Vehicles (UAVs) for rapid data collection in disaster-affected areas, and understand the complexities of combining multiple data sources including vision, language, and radar data. Explore cutting-edge solutions such as generative models for multimodal imagery and explainable interactive models for vision and language perception. Gain insights into the development of a foundation model designed to enhance robot-based multi-modal tasks with minimal labeled data requirements, ultimately contributing to more efficient disaster response and recovery efforts.