Motion-triggered cameras, or “camera trap“, giving everyone from homeowners to park managers an unprecedented view of their local wildlife. While a curious backyard user may be able to identify a creature by eye, larger projects are now collecting thousands or even millions of wildlife images that could take decades to identify manually.
Today, more people than ever before are using AI to identify animals in their images speciesnet. This AI model developed by Google can classify approximately 2,500 animal categories in camera trap images, thanks to conservation partners who provided 65M labeled images to train the model. Originally part of the online platform Wildlife InsightA year ago we released SpeciesNet into the wild open source A tool for others to download, customize, and refine.
Over the past 12 months, research groups around the world have used the open-source SpeciesNet model to observe pumas and ocelots in Colombia, elk and black bears in Idaho, cassowaries and musk rat-kangaroos in Australia, and lions and elephants in Tanzania’s Serengeti National Park. AI models are allowing more people to ask broader questions about wildlife patterns and conservation.
Part of SpeciesNet google earth aiA collection of geospatial tools, datasets, and AI models for deep planetary intelligence. Earth AI empowers communities and nonprofits to address some of the planet’s most urgent needs.
