This is a submission for the Hacktoberfest Open-Source AI Challenge Week 1: Touch Grass
What I Built
TerraQuest is a browser-based, privacy-first web application that gamifies nature exploration. Instead of staring at a screen indoors, TerraQuest encourages you to go outside and interact with the physical world.
Using your device's camera, it acts as an interactive "Scanner" to identify botanical foliage, mineral textures, and other natural items (like pinecones, leaves, and acorns). As you explore, you complete outdoor "Quests," earn XP, unlock badges (like "Woodland Scout" or "Botanist"), and log your discoveries in a personal Journal. It's built for hikers, students, and anyone who needs a fun excuse to step away from their desk and touch grass.
Demo
Live Site: https://terraquestt.netlify.app/
Code
https://github.com/Harshraj979/TerraQuest
How I Built It
TerraQuest is built using pure HTML, CSS, and Vanilla JavaScript. For the AI component, I utilized Transformers.js (@xenova/transformers) to run a local computer vision pipeline directly inside the browser.
Because the inference happens entirely locally on the user's device, there is absolutely no backend server required. It uses navigator.mediaDevices to access the environment-facing camera and processes the video feed in real-time.
Why Does Open Innovation Matter?
Open innovation and open-weight models made this project possible in a way that closed APIs never could.
The entire point of TerraQuest is to go outside into nature—often to places where cellular data is spotty or non-existent (like deep in the woods or on a hiking trail). Because I can run an open-source vision model completely locally via Transformers.js, the app functions off-grid. A closed API would require a constant, strong internet connection, completely ruining the core "touch grass" experience for remote explorers.
Prize Categories
Hacktoberfest Week 1: Touch Grass
Top comments (0)