Slum mapping using citizen science, deep learning and transfer learning methods

Mapping slums in different urban environments across the Globe is extremely important for locating and prioritizing hotspots of urban poor for urgent climate action to combat the impact of climate change. Earth Observation images together with advanced machine learning methods are key solutions used for this task. Yet, these solutions are mostly siloed, and they often lack transferability and scalability and fail to include the opinions of different interest groups. They are mostly top-down, with very little attention given to ground information and interaction with urban communities and stakeholders. Existing top-down approaches should be complemented with bottom-up approaches to produce routinely updated, accurate, and timely slum maps.