JewelAI is a high-end digital jewelry studio that bridges the gap between imagination and reality. By combining generative AI, computer vision, and interactive 3D rendering, our platform allows users to transform images and text descriptions into fully customizable 3D jewelry models with real-time AR try-on capabilities.
Transform any single image of a ring into a high-fidelity 3D model using state-of-the-art neural reconstruction (powered by Wonder3D).
A sophisticated AI concierge (powered by Gemini) that understands your design preferences, detects jewelry components from your uploads, and helps you refine your dream design through natural conversation.
Fine-tune every detail of your jewelry:
- Metal Selection: Gold, Silver, Platinum, Rose Gold, and more.
- Gemstone Studio: Choose stone types (Diamond, Ruby, Emerald, etc.) and precisely scale them.
- Structural Controls: Adjust band width, ring style (Solitaire, Halo, Eternity, Cluster), and profile in real-time.
Experience your creations instantly. Using MediaPipe hand-tracking, the 3D model is projected directly onto your hand in a live camera feed for a virtual "fitting" experience.
Get real-time market value estimations based on your custom design, factoring in metal weight, gemstone quality, and architectural complexity.
- Framework: Next.js 14+ (App Router)
- 3D Engine: Three.js (via React Three Fiber & Drei)
- Computer Vision: MediaPipe (Hand & Pose Tracking)
- State Management: Zustand
- Animations: Tailwind CSS & Framer Motion
- Core: FastAPI (Python)
- AI/ML:
- Gemini 1.5 Pro/Flash: LLM for chatbot and visual analysis.
- Zero123 (optional): For 2D-to-3D generation if configured.
- Wonder3D: Default 2D-to-3D generation.
- TencentARC/InstantMesh: Fallback for stable 3D reconstruction.
- Node.js 18+
- Python 3.10+
- Gemini API Key (Google AI Studio)
- Navigate to the backend directory:
cd backend - Create and activate a virtual environment:
python -m venv venv .\venv\Scripts\Activate # Windows source venv/bin/activate # macOS/Linux
- Install dependencies:
pip install -r requirements.txt
- Create a
.envfile and add yourGEMINI_API_KEY. - Start the server:
uvicorn main:app --reload
- Navigate to the frontend directory:
cd frontend - Install dependencies:
npm install
- Create a
.env.localfile and add:NEXT_PUBLIC_GEMINI_API_KEY=your_key_here NEXT_PUBLIC_API_URL=http://localhost:8000
- Start the development server:
npm run dev
Syrus2026_GDuo/
├── frontend/
│ ├── app/ # Main application routes
│ ├── components/
│ │ ├── viewer/ # 3D Render & AR Logic
│ │ ├── chat/ # Gemini Chat Panel
│ │ └── controls/ # Jewelry Customization UI
│ ├── hooks/ # Hand & Pose tracking hooks
│ └── store/ # Zustand state (Ring & Pricing)
├── backend/
│ ├── routers/ # API endpoints (3D Gen, Chat)
│ ├── services/ # AI & Model Processing services
│ └── models/ # Data schemas
└── outputs/ # Generated 3D assets
- Project Title: AI-Driven 2D to 3D Jewelry Generation with Real-Time Customization
- Repository: https://github.com/CMPN-CODECELL/Syrus2026_GDuo