🧠 ONESHOT3D

ONESHOT3D is a browser-based AI tool that converts any single 2D image into a 3D model using TensorFlow.js and Three.js — all running client-side. It brings real-time, mobile-friendly, interactive 3D model generation from photos.

🌟 Features

• 🖼️ Upload any 2D image (face/object)
• 🤖 AI-based depth estimation via TensorFlow.js
• 🧱 Dynamic 3D model generation using Three.js
• 🔄 Rotate, zoom, and view models in 360°
• 📱 100% browser-based and mobile-ready

🔧 Tech Stack

• HTML, CSS, JavaScript
• TensorFlow.js (AI depth prediction)
• Three.js (3D rendering)
• Mobile-first responsive design

🚀 How to Use

1. Open the app in your browser
2. Upload a photo
3. Wait a few seconds while the model generates
4. Explore the resulting 3D model interactively

📁 Folder Structure

ONESHOT3D/ ├── index.html ├── style.css ├── script.js ├── tf_model/ # TensorFlow.js model files ├── assets/ # Sample images or textures └── output_model/ # Auto-generated 3D models

🧠 AI Model

• Uses a custom-trained TensorFlow.js model for depth prediction
• Optionally integrates with MediaPipe for face detection
• Outputs vertices and depth maps dynamically

🎯 Use Cases

• Avatar generation
• Face-to-3D modeling
• AR/VR applications
• Mobile AI art tools

👤 Author

Naga Sasank (NSSTUDIO-html)
🔗 GitHub: NSSTUDIO-html

📝 License

This project is open-source under MIT License.