Google unveils SpeciesNet – AI Model for wildlife identification
Google has open-sourced SpeciesNet, an AI model designed for automatically identifying animals in camera trap images, Kazinform News Agency correspondent reports, citing TechCurnch.
Biologists and ecologists widely use camera traps to monitor wildlife, but manually processing thousands of images takes time. To streamline this work, Google developed Wildlife Insights, a platform powered by SpeciesNet that helps researchers analyze images more efficiently.
According to Google, SpeciesNet was trained on 65 million images, including those provided by organizations like the Smithsonian Conservation Biology Institute, the Wildlife Conservation Society, the North Carolina Museum of Natural Sciences, and the Zoological Society of London. The model can identify over 2,000 categories, including animal species, broader taxa, and even non-biological objects like vehicles.
By open-sourcing SpeciesNet, Google aims to support researchers, developers, and startups in ecology, helping to expand wildlife monitoring and improve conservation efforts. The model is available on GitHub under the Apache 2.0 license for both scientific and commercial use.
Earlier, Kazinform News Agency reported that a snow leopard, listed as endangered, had once again been captured on camera in the Altyn-Emel National Park.