Google open-sources SpeciesNet AI model to help with wildlife identification

On March 4, Google announced the open-sourcing of an artificial intelligence model called SpeciesNet, which is designed to identify animal species by analyzing photos captured by cameras. To address this problem, Google launched the Wildlife Insights project six years ago as part of its Google Earth Outreach philanthropy program. The platform allows researchers to share, identify and analyze wildlife images online to speed up the processing of camera-capture data.The SpeciesNet model is at the heart of the platform, which Google says is trained on 65 million publicly available images, as well as data from the Smithsonian Institution's Institute for Conservation Biology, the Wildlife Conservation Society, the North Carolina Museum of Nature and Science and the Zoological Society of London, among others. SpeciesNet can classify images into more than 2,000 labels, covering animal species, categories like "mammal" or "feline," and non-animal objects such as "vehicle.