For many people, the climate crisis feels like it only belongs to faraway polar bears, melting ice caps, and people living on low-lying islands like the Maldives.
In a bid to bring the impending issue of climate change closer to home, researchers from the Mila Quebec Artificial Intelligence Institute have created an AI-powered platform that literally shows users what their homes will look like because of climate change-induced natural disasters.
Due to the melting ice caps and rising sea-levels, the biggest disaster is flooding.
RELATED: 4 IMPORTANT FACTS ABOUT CLIMATE CHANGE EVERYONE MUST KNOW
Climate change and Mila
Aside from those who deny climate change is a real issue, many of those who do believe it is happening do not necessarily feel a need for dramatic alterations to their day-to-day lives. The issues simply seem too far away from home.
To try and remedy this, the Mila team of researchers decided to create their platform, where people can upload images of streets they know, then an AI-driven algorithm shows them what these would look like after being flooded.
The idea is to allow people to make more informed decisions around climate change. Should they keep living the way they do and most likely keep fuelling the crisis, or should they fight for their earth's future?
Victor Schmidt, a PhD candidate at Mila and head author of a paper that outlined their work, said "there are so many consequences of climate change. It’s going to impact everyone. We want to help people better understand that and help them engage in actually taking action."
The Mila visualization platform
Currently, the Mila visualization platform only shows flooded streets, but it aims to show images of other climate-related events such as wildfires. These images are hard to depict so the team require more time before effectively putting the option out to the public.
The researchers' method may be quite a strong one, as Schmidt continued "We feel that showing people the potential consequences of climate change in their neighborhoods is a good way of making climate change more personal and less distant."
The Mila team used an image-to-image translation algorithm to transform people's uploaded images from Google Street View into ones that displayed the after-effects of flooding.
This was done using a generative adversarial network (GAN) that trained the system.
The ways GAN works is by pitting two algorithms against each other, one that generates an image, and the other that tries to decipher whether its real or fake. This allows the images to look more real as the first algorithm tries harder and harder to create believable and realistic images to trick the second one.
The main issue the Mila team faces at the moment is having enough aftermath disaster-stricken images that they can use to train the algorithm. So the group launched ClimatePix earlier this year for the public to upload their images.
Schmidt and his team are not climate scientists, nor is their platform scientifically backed. They see themselves as communicators who help people become more informed and aware of the climate change situation.