© 2021 Strange Loop
Each year, 11 billion dollars is spent cleaning up litter. And that’s just in the US. Almost all of this money is spent without knowing where the litter is located, nor its composition. What are the objects, materials, and brands we see thrown on the ground? All around the world, cities and corporations are increasingly being called to help solve the global issue of litter in the environment.
8 years ago, Litterati began as an Instagram picture of trash and has now grown into one of the world’s largest open litter data sources in the world encompassing over 10 million images and locations of trash from 185 countries.
We use computer vision and machine learning to recognize and label each image with its object type, material and brand. We’re also learning information like the weight and volume of the trash, and how long it's been decomposing on the ground. We're also scaling our data collection to include new sources of data like street cameras and drone footage to further our data footprint.
Our solution centers on serverless technology and step functions allowing for customization of our recognition capabilities to support different models for different scenarios and environments. We're using the data to not only inform those in power about where trash is, but also using world-class scientific methodology to leverage our data to inform policy, influence packaging, and inspire personal responsibility.
Sean Doherty is the CTO at Litterati, a company focused on bringing data-driven change to the world’s litter problem through collection and analysis of pictures of trash from all over the world. He is passionate about mentorship and bringing equity to software through training, advocacy and support. All of his free time is enjoyed with his family and five children who are all the hobby he'll ever need.