README.md

metadata

title: Watermark Remover
emoji: 🖼️
colorFrom: blue
colorTo: purple
sdk: docker
app_port: 7860
pinned: false

Watermark / Text Remover

AI-powered tool for removing watermarks, text, and unwanted objects from images.

Features

• Watermark / Text Removal — LaMa inpainting model
• Object Removal — free-draw mask + LaMa inpainting
• Background Removal — BiRefNet segmentation
• Image Upscaling — Real-ESRGAN 4×
• Face Restoration — CodeFormer + ArcFace identity guard
• Face Zoom — per-face crop, upscale, and restore

Usage

Upload an image, select a feature from the sidebar, and follow the on-screen instructions. All processing runs on CPU — large images may take 10–30 seconds per operation.

Local Development

# Backend (Python 3.11+)
cd backend
pip install -r ../requirements.txt
uvicorn app.main:app --reload --port 8000

# Frontend (Node 18+)
cd frontend
npm install
npm run dev

Open http://localhost:5173 in your browser.

Model Weights

Model	Source	Size
LaMa (lama.pth)	Git LFS (committed to repo)	~197 MB
Real-ESRGAN (RealESRGAN_x4plus.pth)	Downloaded at build time	~64 MB
CodeFormer (codeformer.pth)	Downloaded at build time	~360 MB
BiRefNet	Downloaded at build time (HF Hub)	~424 MB
RetinaFace + ParseNet	Downloaded at build time (facexlib)	~185 MB
ArcFace IR-SE50	Downloaded at build time (facexlib)	~167 MB
PaddleOCR	Downloaded at build time	~200 MB