Despite the incredible power of geospatial information, only a small group of people historically have been able to leverage it for informed decision-making. This situation has been exacerbated by the exponential explosion of data from satellites, drones, and Internet of Things sensors, exposing how traditional GIS and remote sensing techniques are not equipped to handle this tsunami of spatial data.
Rapid advancements in computer vision and deep learning techniques promise to change the game, enabling more people to tap into the power of geospatial. In the next few years, we could see a radical transformation of remote sensing as these new machine learning methods are applied to earth observation data. Indeed, a fully automated approach enabled by new technology is arguably the only chance the geospatial industry has to enable meaningful decisions from vast amounts of imagery and geospatial information.
How can we as an industry capitalize on this massive opportunity? This talk presents a vision for building towards a "Queryable Earth" – where satellite imagery activated by deep learning and computer vision techniques can unlock the power of geospatial information and put this power into the hands of more people.