microsoft/Biodiversity
Microsoft AI for Good Lab — Biodiversity research hub. Open-source AI models, edge devices, and tools for biodiversity monitoring and conservation. Your source for MegaDetector, SPARROW, PytorchWildlife, Bioacoustics, and more.
What it solves
This project provides a suite of AI tools designed for biodiversity monitoring and wildlife conservation. It addresses the challenges of processing vast amounts of data from camera traps, bioacoustic sensors, and aerial imagery to identify and monitor species of animals, people, and vehicles.
How it works
This repository acts as a central hub for several specialized AI projects:
- MegaDetector: An AI model for detecting animals, people, and vehicles in camera-trap imagery.
- MegaDetector-Acoustic: A bioacoustic AI for audio classification and species identification from sound.
- MegaDetector-Overhead: A tool for point-based wildlife localization from aerial views.
- MegaDetector-Sonar: A system for processing and feature detection in sidescan sonar imagery.
- MegaDetector-Classifier: A framework for fine-tuning species classification models to specific datasets and regions.
- PyTorch-Wildlife: A collaborative deep learning framework and model zoo for conservation AI.
- SPARROW: A solar-powered AI edge device for remote field deployments.
Who it’s for
Conservationists, researchers, and NGOs working in wildlife monitoring and biodiversity conservation using various sensor data types.
Highlights
- Multimodal Monitoring: Supports imagery (camera traps, overhead, sonar) and audio (bioacoustics).
- Edge Deployment: Includes hardware for remote field deployments via SPARROW.
- Unified Framework: PyTorch-Wildlife provides a collaborative model zoo for conservation AI.
- Wide Adoption: Used by national parks, research universities, and NGOs globally.