Google launches SpeciesNet, a fun AI tool for spotting wildlife

Google has recently open sourced an AI model called SpeciesNet, which is designed to identify animal species by analyzing photos from camera traps. Camera traps are digital cameras connected to infrared sensors that researchers around the world use to study wildlife populations. However, the sheer volume of data generated by these traps can take days to weeks to sift through.

In an effort to assist researchers, Google launched Wildlife Insights about six years ago. This initiative, part of the Google Earth Outreach philanthropy program, provides a platform for researchers to share, identify, and analyze wildlife images online, helping to speed up camera trap data analysis. Many of the analysis tools on Wildlife Insights are powered by SpeciesNet, which Google claims was trained on over 65 million publicly available images and images from various organizations.

SpeciesNet can classify images into over 2,000 labels, covering animal species, taxa like “mammalian” or “Felidae,” and even non-animal objects like “vehicle.” Google believes that the release of the SpeciesNet AI model will enable developers, academics, and biodiversity-related startups to scale monitoring of biodiversity in natural areas. The AI model is available on GitHub under an Apache 2.0 license, allowing for commercial use with minimal restrictions.

It’s important to note that Google is not the only company offering open source tools for automating the analysis of camera trap images. Microsoft’s AI for Good Lab also maintains PyTorch Wildlife, an AI framework with pre-trained models specifically tailored for animal detection and classification.