ColorMatch Studio

AI-powered personal color analysis. Upload a portrait, get flattering clothing color & pattern recommendations with visual try-on previews.

Stack

• Frontend: Next.js 14 + TypeScript + Tailwind CSS
• Backend: FastAPI + Python + OpenCV + MediaPipe

Quick Start

# Backend
cd backend
python3.11 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
uvicorn main:app --reload --port 8000

# Frontend
cd frontend
npm install
npm run dev

How It Works

1. Face Detection — MediaPipe Face Mesh extracts 468 landmarks
2. Skin Sampling — Samples from cheek, forehead, chin zones; removes outliers
3. Color Features — Computes CIELAB, HSV, RGB stats, redness/yellowness, contrast
4. Classification — Rule-based scorer predicts undertone, depth, chroma, contrast → seasonal bucket
5. Palette Ranking — 80+ colors scored against profile across 4 weighted criteria
6. Preview Rendering — Segments torso, applies HSV-shift recolor and procedural patterns

API

Method	Path	Description
POST	/api/analyze	Upload image, run analysis
GET	/api/result/{id}	Get stored result
POST	/api/override	Override undertone/contrast
POST	/api/generate-preview	Single recolor/pattern preview
POST	/api/generate-gallery	Full preview gallery
GET	/demo-result	Demo seed data

License

MIT