GEOspatial Artificial Intelligence (GEOAI) is a research group located at the National Center for Remote Sensing - CNRS Lebanon. GEOAI harnesses the power of artificial intelligence to unlock the potential of satellite data by focusing on AI-assisted mapping spanning various applications, including geospatial Earth observation, urban analytics, transportation, and features extraction from aerial imagery. We develop tools integrating deep learning techniques to automate the process of urban features extraction and collect crowd-sourcing data from various sources. Evidence gained from our models and data analysis allows for a robust humanitarian response and provides policymakers and key stakeholders with insights to design tailored regulations and safety countermeasures for urban social good.
GEOAI group provides a unique opportunity for geospatial training in a professional research environment in Lebanon. We have hosted more than 40 students at both undergraduate and graduate levels since 2016.
The first comprehensive Lebanese Building Footprints autonomously generated using Deep Learning.
Solar Potential Assessment using Multi-Class Buildings Segmentation from Aerial Images
The dataset is a collection of RGB images of Beirut buildings taken from satellite images. Each building is annotated as residential or not. Dataset can be accessed here.