Emoji Mask

Hide faces with emojis before you post — entirely in your browser.

Emoji Mask is a single-file web app that detects faces in photos and covers them with emojis. Photos never leave your device. Exported images are re-encoded, so EXIF metadata (including GPS location) is stripped automatically.

Features

• Automatic face detection — powered by face-api.js (SSD MobileNet v1), with a tiled second pass on large images
• Privacy-first — all processing runs locally; no uploads, no server
• Works offline — model weights are embedded in index.html
• Emoji themes — smiles, cats, animals, fruits, space, sweets, ocean, spooky, hearts, or one emoji for every face
• Manual editor — move, resize, add, or delete emoji overlays; swipe or use arrow keys to move between photos
• Batch export — download individual images or everything as a .zip

Quick start

No install or build step required.

Option 1: Open the file

Download or clone this repo, then open index.html in a modern browser (Chrome, Edge, Firefox, or Safari).

The file is ~8 MB because it bundles the face detector and model weights for offline use. Make sure it finishes downloading before opening.

Option 2: Local server (recommended for development)

git clone https://github.com/KULcoder/emojiBlur.git
cd emojiBlur
python3 -m http.server 8080

Then visit http://localhost:8080.

Deploy to GitHub Pages

This repo includes a GitHub Actions workflow that publishes the site on every push to main.

1. Push this repository to GitHub.

2. Go to Settings → Pages → Build and deployment.

3. Set Source to GitHub Actions.

4. After the workflow runs, the site will be live at:

   https://KULcoder.github.io/emojiBlur/

You can also deploy manually: enable Pages with the main branch and / (root) as the source — index.html at the repo root is all you need.

How it works

1. Drop or select one or more photos (JPEG, PNG, WebP).
2. The detector finds faces and places emoji overlays sized to each face box.
3. Tap Edit on any photo to fine-tune placements, change emojis, or add masks manually.
4. Save individual images or Download all (.zip).

If automatic detection is unavailable (blocked CDN, ad blocker, etc.), you can still add photos and place emojis manually.

Browser support

Browser	Support
Chrome / Edge	✅ Full
Firefox	✅ Full
Safari	✅ Full (iOS 15+)

HEIC/HEIF images may not decode in all browsers — export as JPEG first if needed.

Project structure

emojiBlur/
├── index.html          # Entire app (UI, logic, face-api.js, JSZip, model weights)
├── README.md
├── LICENSE
└── .github/workflows/  # GitHub Pages deployment

There is no build pipeline. Everything lives in one self-contained HTML file.

Credits

• face-api.js — face detection (MIT)
• JSZip — ZIP export (MIT/GPL dual license)
• Nunito — UI font (Google Fonts)

License

MIT — see LICENSE.