BirdLens AI: Identify Birds Locally, Explore Nature Offline
BirdLens AI — An Open-Source AI Bird Identification Companion This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass . What I Built I built BirdLens AI , a browser-based bird-identification companion for beginner birdwatchers, curious walkers, and anyone who wants to learn more about the wildlife around them. My approach to “Touch Grass” is simple: use technology…
BirdLens AI is an open-source AI bird identification tool designed for beginner birdwatchers and curious walkers. The web-based application allows users to upload bird photos and receive suggested identifications using an open-weight model. Users can review the top five predictions, access information about the species, and save their discoveries in a local journal.
The app also encourages users to take part in observation challenges to explore their surroundings more closely. The project was built using a variety of technologies, including React and TypeScript for the interface and application logic, Vite for development and production builds, Tailwind CSS for styling, ONNX Runtime Web for browser-based inference, IndexedDB for local storage, and a Service Worker for offline support.
The core of BirdLens AI is an open-weight model called chriamue/bird-species-classifier, which is an EfficientNet-B2 model with 525 output classes. The model's weights are loaded in ONNX format and executed through ONNX Runtime Web. By using an open-weight model and running inference locally in the browser, BirdLens AI enables users to identify birds without the need for an internet connection after the necessary assets have been cached.
The application is deployed on Vercel, and all necessary assets, including the model, are included in the deployment. BirdLens AI promotes responsible birdwatching by encouraging users to observe birds from a respectful distance and not to approach or disturb wildlife.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
