microsoft/Pytorch-Wildlife

microsoft/Pytorch-Wildlife

Description: PyTorch-Wildlife — The Microsoft collaborative deep learning framework for conservation. Multi-task computer vision and bioacoustic models for camera-trap detection, species classification, and biodiversity monitoring. Maintained by Microsoft AI for Good Lab.

Language: Python

License: MIT

Stars: 6

Forks: 0

Open issues: 4

Created: 2026-05-06T02:06:49Z

Pushed: 2026-06-07T00:49:25Z

Default branch: main

Fork: no

Archived: no

README: !PyTorch-Wildlife — open-source AI framework for wildlife monitoring from the Microsoft AI for Good Lab

PyTorch-Wildlife

Unified open-source AI framework for wildlife monitoring and conservation. Microsoft AI for Good Lab — camera-trap detection, species classification, bioacoustic analysis, and more.

PyTorch-Wildlife is the collaborative deep learning framework that powers the Microsoft AI for Good Lab's biodiversity work. It hosts detection models, species classifiers, and the tools needed to run them — from single-image inference to large-scale batch processing.

MegaDetector, the most widely used camera-trap detection model in conservation, is invoked through PyTorch-Wildlife. So are the species classifiers for Amazon Rainforest, Snapshot Serengeti, and European ecosystems.

Quick Start

pip install PytorchWildlife

import numpy as np
from PytorchWildlife.models import detection as pw_detection
from PytorchWildlife.models import classification as pw_classification

# Detection — weights download automatically
detection_model = pw_detection.MegaDetectorV6()
detection_result = detection_model.single_image_detection("path/to/image.jpg")

# Classification
classification_model = pw_classification.AI4GAmazonRainforest()
classification_result = classification_model.single_image_classification("path/to/image.jpg")

Try without installing anything:

• Hugging Face demo — upload images in your browser
• Google Colab notebook — free cloud GPU

Available Models

Detection

| Model | Architecture | Description | |---|---|---| | MegaDetectorV6 | YOLOv10 / YOLOv9 / RT-DETR | Animal detection in camera-trap images | | MegaDetectorV5 | YOLOv5 | Previous generation, widely deployed | | DeepfauneDetector | YOLOv8 | European ecosystem detection | | HerdNet | CNN localization | Point-based detection for aerial imagery |

Classification

| Model | Description | |---|---| | AI4GAmazonRainforest | Species classification for Amazon Rainforest | | AI4GSnapshotSerengeti | Species classification for African savanna | | AI4GOpossum | Opossum vs. non-opossum classifier | | DeepfauneClassifier | European ecosystem species classifier | | DFNE | Deepfaune fine-tuned for Northeastern North America |

See the Model Zoo for full details, performance benchmarks, and version history.

Part of the Biodiversity Ecosystem

PyTorch-Wildlife is part of the larger open-source ecosystem from the Microsoft AI for Good Lab:

| Repo | Purpose | |---|---| | microsoft/Biodiversity | The umbrella repository — documentation hub for the AI for Good Lab's biodiversity work | | microsoft/Pytorch-Wildlife | This repo — the unified deep learning framework | | microsoft/MegaDetector | Animal detection in camera-trap imagery (documentation) | | microsoft/SPARROW | Solar-Powered Acoustic and Remote Recording Observation Watch — AI-enabled edge device | | microsoft/MegaDetector-Acoustic | Bioacoustic models for audio-based wildlife monitoring | | microsoft/MegaDetector-Classifier | Camera-trap species classification fine-tuning — adapt classifiers to your own datasets and geographic regions | | microsoft/MegaDetector-Overhead | Point-based detection for overhead and aerial imagery | | SPARROW Studio | Desktop application for running all models with a graphical interface |