Skip to content

Crusherbolt/ewigen

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

8 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

⏳ Ewigen - Timeless Moments

Live Demo License Python Docker

Preserve your most precious moments forever through immersive virtual reality



🌟 Project Overview

Ewigen (German for "Timeless") is a cutting-edge platform that transforms multi-angle video recordings into fully interactive 3D virtual environments. Capture real-world spaces and moments, then relive them in immersive VR with AI-powered interactive characters.

Two Revolutionary Use Cases

1. Humanoid Robot Training Datasets (B2B) πŸ€–

Generate high-fidelity 3D environments and human motion datasets for training humanoid robots and autonomous systems. Export data in industry-standard formats (USD, ROS, URDF) compatible with NVIDIA Omniverse, Isaac Sim, and major robotics frameworks.

Value Proposition:

  • Synthetic training data generation
  • Realistic human motion capture
  • Multi-modal sensor simulation
  • Scalable dataset production

2. Immersive Memory Recreation (B2C) πŸ’

Transform special momentsβ€”weddings, family gatherings, concertsβ€”into interactive VR experiences. Walk through reconstructed spaces, interact with AI-powered avatars of people, and relive memories as if you were there.

Value Proposition:

  • Preserve precious moments forever
  • Interact with loved ones in VR
  • Relive experiences from any angle
  • Share memories with future generations

✨ Key Features

  • πŸ“Ή Multi-View 3D Reconstruction: Convert multiple video angles into photorealistic 3D environments using Neural Radiance Fields (NeRF) and Gaussian Splatting
  • πŸ‘€ Human Avatar Generation: Automatically detect, track, and create realistic 3D avatars from video footage
  • 🎭 Motion Capture: Extract natural human movements and gestures for animation
  • πŸ—£οΈ Voice Cloning: Replicate voices for authentic character interactions
  • πŸ€– AI Conversations: Engage in natural conversations with virtual characters powered by advanced language models
  • πŸ₯½ VR Experience: Explore reconstructed environments in immersive virtual reality
  • πŸ“Š Dataset Export: Generate training datasets for robotics and AI research
  • ☁️ Cloud Processing: Scalable infrastructure for processing large video collections

πŸ—οΈ Architecture

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                     Frontend (Next.js)                       β”‚
β”‚              Web Dashboard + Project Management              β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                              β”‚
                              β–Ό
β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
β”‚                   Backend API (FastAPI)                      β”‚
β”‚         Authentication β”‚ Projects β”‚ Processing Queue         β”‚
β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                              β”‚
                β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”Όβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
                β–Ό             β–Ό             β–Ό
    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”  β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
    β”‚ 3D Recon     β”‚  β”‚ Character    β”‚  β”‚ AI Services  β”‚
    β”‚ - NeRF       β”‚  β”‚ - Detection  β”‚  β”‚ - Voice      β”‚
    β”‚ - Gaussian   β”‚  β”‚ - Avatars    β”‚  β”‚ - Speech     β”‚
    β”‚ - COLMAP     β”‚  β”‚ - Animation  β”‚  β”‚ - LLM        β”‚
    β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜  β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜
                              β”‚
                              β–Ό
                    β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
                    β”‚  VR Application  β”‚
                    β”‚     (Unity)      β”‚
                    β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜

πŸ“Š View Full Interactive Architecture


πŸš€ Quick Start

Prerequisites

  • Python 3.10+
  • Node.js 18+
  • Docker & Docker Compose
  • NVIDIA GPU (recommended: RTX 3090 or better with 24GB+ VRAM)
  • CUDA 11.8+
  • 16GB+ RAM (32GB recommended)

Installation

  1. Clone the repository
git clone https://github.com/YOUR_USERNAME/ewigen.git
cd ewigen
  1. Set up environment variables
cp .env.example .env
# Edit .env with your configuration
  1. Run with Docker (Recommended)
# Make deployment script executable (Linux/Mac)
chmod +x deploy.sh

# Or on Windows PowerShell
# No need to chmod

# Start all services
./deploy.sh local
  1. Access the application

πŸ“– Usage

1. Upload Videos

Upload multiple video recordings of your environment from different angles through the web dashboard or API.

2. Process Videos

The platform automatically:

  • Extracts frames and synchronizes videos
  • Calibrates cameras using COLMAP
  • Reconstructs 3D environment with NeRF
  • Detects and tracks people
  • Generates 3D avatars
  • Clones voices from audio
  • Creates interactive AI characters

3. Export Results

For Robot Training:

# Export as USD format (NVIDIA Omniverse)
curl -X POST "http://localhost:8000/api/v1/projects/{id}/export" \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -d '{"format": "usd", "include_annotations": true}'

For VR Experience:

  • Download the Unity VR application
  • Load your reconstructed environment
  • Put on your VR headset and explore

πŸ› οΈ Technology Stack

Backend

  • FastAPI: High-performance async Python web framework
  • PostgreSQL: Relational database for metadata
  • Redis: Caching and message broker
  • Celery: Distributed task queue for video processing
  • MinIO: S3-compatible object storage

Machine Learning

  • PyTorch: Deep learning framework
  • NeRF (Instant-NGP): 3D scene reconstruction
  • Gaussian Splatting: Real-time rendering (60+ FPS)
  • COLMAP: Structure from Motion
  • YOLOv8: Human detection
  • MediaPipe: Pose estimation
  • SMPL-X: Parametric human body model
  • Whisper: Speech recognition
  • Coqui TTS: Voice synthesis
  • GPT-4/Claude: Conversational AI

Frontend

  • Next.js 14: React framework with App Router
  • TypeScript: Type-safe JavaScript
  • Tailwind CSS: Utility-first CSS framework
  • Three.js: 3D visualization

VR

  • Unity: VR application development
  • XR Interaction Toolkit: VR interactions
  • Universal Render Pipeline: High-quality rendering

πŸ“Š Dataset Formats

The platform exports datasets in multiple formats:

  • USD/USDA: NVIDIA Omniverse, Pixar formats
  • ROS Bags: Robot Operating System
  • URDF: Unified Robot Description Format
  • FBX/OBJ: Standard 3D model formats
  • JSON: Annotations and metadata

πŸ“ˆ Performance Metrics

  • Processing Time: 4-7 hours for 50 minutes of video
  • VR Frame Rate: 60+ FPS with Gaussian Splatting
  • Storage: 10-25 GB per project
  • Scalability: Horizontal scaling with multiple GPU workers

🎯 Use Cases

Robotics & AI Research

  • Train humanoid robots in realistic environments
  • Generate synthetic training data
  • Test navigation algorithms
  • Simulate human-robot interactions

Entertainment & Events

  • Preserve wedding memories in VR
  • Create virtual concert experiences
  • Archive family gatherings
  • Build interactive museum exhibits

Real Estate & Architecture

  • Virtual property tours
  • Construction progress documentation
  • Interior design visualization
  • Historical building preservation

Education & Training

  • Immersive learning environments
  • Medical procedure training
  • Safety training simulations
  • Historical event recreation

πŸ”§ Configuration

Key configuration options in .env:

# Database
DATABASE_URL=postgresql://user:pass@localhost:5432/ewigen

# Storage
MINIO_ENDPOINT=localhost:9000
MINIO_ACCESS_KEY=minioadmin
MINIO_SECRET_KEY=minioadmin

# AI Services
OPENROUTER_API_KEY=your_key_here
ELEVENLABS_API_KEY=your_key_here

# Processing
MAX_WORKERS=4
GPU_MEMORY_FRACTION=0.8

πŸ§ͺ Testing

# Backend tests
cd backend
pytest tests/ -v

# Frontend tests
cd frontend
npm test

# ML pipeline tests
cd ml
pytest tests/ -v

πŸ“¦ Deployment

Docker Deployment

./deploy.sh production

Cloud Deployment

Supports deployment to:

  • AWS: EC2, ECS, S3, RDS
  • Google Cloud: Compute Engine, Cloud Storage, Cloud SQL
  • Azure: Virtual Machines, Blob Storage, Azure Database

See CLOUD_DEPLOYMENT_GUIDE.md for detailed instructions.


πŸ“ Documentation


🀝 Contributing

We welcome contributions! Please follow these steps:

  1. Fork the repository
  2. Create a 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

πŸ“„ License

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


πŸ†˜ Support

For issues and questions:

  • GitHub Issues: Report bugs and request features
  • Email: support@ewigen.com
  • Documentation: Full documentation in /docs

🎯 Roadmap

  • Real-time collaborative VR experiences
  • Mobile app for video capture
  • Advanced physics simulation
  • Multi-language support
  • Marketplace for datasets
  • Integration with major robotics platforms

πŸ’‘ Why "Ewigen"?

Ewigen is German for "Eternal" - perfectly capturing our mission to preserve timeless moments. The name conveys:

  • Timelessness: Moments that transcend time
  • Sophistication: European elegance and precision
  • Technology: Advanced engineering heritage
  • Emotion: Deep connection to preservation

Alternative meanings:

  • Everlasting World Immersive Generation Engine Network
  • Represents the timeless nature of digitally preserved memories

πŸ† Built With

  • ❀️ Passion for preserving memories
  • 🧠 Cutting-edge AI and ML research
  • 🎨 Beautiful design and UX
  • ⚑ High-performance engineering
  • 🌍 Global collaboration

⏳ Ewigen - Preserve timeless moments forever

Where the past meets the future in immersive virtual reality

About

Synthetic dataset generation for humanoid robotics. Multi-view 3D reconstruction, human motion capture & scene annotation. USD | ROS | URDF export. ( NVIDIA Omniverse compatible.)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors