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.

Animal Detection using YOLOv5

Wild Animals Detection and Alert System

Animal Detection in Man-made Environments using Deep Learning

Animal Detection and Classification using YOLO

A U-Net-based deep learning ensemble model for wildebeest-sized animal detection from satellite imagery. A version used for the paper accepted by Nature Communications: "Deep learning enables satellite-based monitoring of large populations of terrestrial mammals across heterogeneous landscape" (https://rdcu.be/dc8bU).

Detection of Animals in camera trapped images using RetinaNet in pytorch.

Animals object detection such as deer, horse, and rabbit in diverse settings using YOLOv5

🐬 Code accompanying the article "Weakly Supervised Detection of Marine Animals in High Resolution Aerial Images"

Closed-loop feedback optogenetic system

Image Processing and Deep Learning algorithm to detect leopards from a live camera feed.

YOLOv11-based object detection for civilians, rescuers, and animals in disaster scenarios. Real-time support for emergency response in floods and structural collapses.

Agri-Pal is the simplest solution to aid a farmer in Agriculture - Crop and Poultry Farming. Agri-Pal is a simple Plug n Play device ensuring Disease Detection and Animal Breach Detection.

This Python-based code that utilizes OpenCV's DNN module with MobileNetSSD to detect animals in the farmland.The code provides a GUI using Tkinter, allowing users to select a video file and start the animal detection process. When an animal is detected, an alert is triggered with a siren sound.

Homebridge plugin for local animal, person, package and vehicle detection from camera/video streams, exposed as HomeKit motion sensors.

An application which uses the camera to detect and identify animals, sounding an alert after their detection and emailing the owner about it

YOLOv8 Animal Detection for Embedded Systems. 97% test accuracy in just 400kb (about the same size as the photos it classifies or 1 second of video). Various quantization, pruning, and distillation techniques for vision models are explored.