Google open-sources SpeciesNet to enhance wildlife monitoring

Google has open-sourced SpeciesNet, an advanced AI model designed to identify animal species from images captured by camera traps. The model, aimed at supporting researchers in analysing vast amounts of wildlife data, is now available on GitHub under an Apache 2.0 license, enabling commercial and academic use with minimal restrictions.

Camera traps, equipped with infrared sensors, are widely used by researchers to monitor wildlife populations. However, sorting through the massive volume of images generated can take days or even weeks. To address this challenge, Google introduced Wildlife Insights six years ago as part of its Google Earth Outreach initiative. This platform enables researchers to share, classify, and analyse wildlife images efficiently.

SpeciesNet plays a crucial role in Wildlife Insights, leveraging AI to automate image analysis. Trained on more than 65 million publicly available images, as well as datasets from organisations like the Smithsonian Conservation Biology Institute, the Wildlife Conservation Society, and the Zoological Society of London, SpeciesNet is capable of classifying images into over 2,000 categories. These include specific animal species, broader taxonomic groups like “mammalian” or “Felidae,” and even non-animal objects such as vehicles.

According to Google, the release of SpeciesNet will empower developers, academics, and biodiversity-focused startups to scale up monitoring efforts in natural environments. 

Google’s move is expected to accelerate biodiversity research and conservation efforts by providing a robust tool for automated image classification. However, SpeciesNet is not the only AI-driven open-source solution in this domain. Microsoft’s AI for Good Lab has also developed PyTorch Wildlife, an AI framework offering pre-trained models for detecting and classifying animals in camera trap images.

With increasing interest in AI-powered conservation tools, the release of SpeciesNet marks a significant step in using machine learning to support global biodiversity monitoring efforts.