Google uses AI to help build cities

Google has just released an AI-powered tool which can track building changes called the Open Buildings 2.5D Temporal Dataset.

The dataset spans a 32 million square mile area across Africa, Latin America, and South and Southeast Asia. The tracking in changes includes estimates of building presence, counts, and heights, and works between the years 2016 to 2023.

To plan for population growth, respond to crises, and understand urbanization’s impact, the technology giant says data about buildings and infrastructure is needed.

This will become more important as time goes on as the world’s urban population is expected to increase by 2.5 billion by 2050, with almost 90% of that growth occurring in cities across Asia and Africa.

It was in 2021 when the company first started its quest to provide more data about these regions, with the Open Buildings dataset launching during that year. This increased the number of publicly mapped buildings in Africa which was then expanded into Latin America, the Caribbean, and South and Southeast Asia.

Announcing the Open Buildings 2.5D Temporal Dataset, which expands the Open Buildings effort by providing data about building presence over time and height data across the Global South from 2016 to 2023. Learn more at: https://t.co/vNihs35Ulj pic.twitter.com/yp2ptY2kPv

— Google AI (@GoogleAI) September 19, 2024

The data has now been improved upon even further with the news of the 2.5D Temporal addition. To detect the buildings, the Sentinel-2 public satellite was used for the imagery and was made available to the company by the European Space Agency.

The Sentinel-2 allows for every point on Earth to be captured roughly every five days, with each pixel on the ground being a 10 m square.

Since its creation, the 2.5d dataset has been shared with Google partners who can then leverage it within their work. This includes WorldPop who create estimates of global populations, UN Habitat who deal with urban sustainability, and Sunbird AI which has used the data for urban planning and rural electrification.

The researchers hope that government agencies will be able to “gain valuable insights into urban growth patterns to inform planning decisions and allocate resources effectively.”

There’s also a belief it could be of use to humanitarian organizations to “quickly assess the extent of built-up areas in disaster-stricken regions, enabling targeted aid delivery.” As well as for researchers to “track development trends, study the impact of urbanization on the environment, and model future scenarios with greater accuracy.”

Featured Image: Via Google Research Blog

The post Google uses AI to help build cities appeared first on ReadWrite.