Ecological Digital Twin Platform for Landscape Design

This is a 3D visualization platform for landscape designers working with photorealistic digital twins. It integrates 3D Gaussian Splatting with ecological modeling to support native plant-based landscape design decisions.

See live demo at https://digitaltwin.ecodash.ai.

• 3D Gaussian Splatting uses advanced neural network reconstruction for lifelike rendering
• Geospatially accurate capture from drone and ground surveys
• Minimal visual difference between digital interface and reality

• Visualize landscape design proposals through interactive polygon overlays on photorealistic digital twins
• Zero-lag interaction for immersive design experience

• Soil Chemistry: pH, nutrient levels (NPK), and composition analysis
• Microclimate Mapping: Solar exposure, moisture gradients, and wind patterns
• Environmental Risk: Drought, flood, and extreme weather vulnerability assessment

Requirements: Python 3.7+, Node.js 18+ and npm

git clone https://github.com/legel/digitaltwin.git

cd digitaltwin

pip install -r requirements.txt

SuperSplat is fully integrated as local files in the supersplat-build/ directory:

# One-command setup for SuperSplat (installs dependencies, builds, and deploys)

npm run setup:supersplat

Or run individual steps:

# Install SuperSplat dependencies

npm run install:supersplat



# Build SuperSplat

npm run build:supersplat  



# Deploy built files to serving directory

npm run deploy:supersplat

Fully Decoupled: All SuperSplat source code is maintained locally in the supersplat-build/ directory. The SuperSplat application files are built in supersplat-build/dist/ and copied to supersplat/ where they are served by the digitaltwin server. No external repositories or complex dependency management required.

The digitaltwin server.py serves SuperSplat files through its generic static file route:

# Serve static files (CSS, JS, images) - includes SuperSplat files in supersplat/ directory

@app.route('/<path:path>')

def serve_static(path):

    return send_from_directory('.', path)

This automatically serves SuperSplat files from the supersplat/ directory, allowing the digitaltwin interface to load SuperSplat in iframe mode at /supersplat/index.html.

python server.py

# Open http://localhost:5001

Development Environment: Direct file editing with browser refresh. No complex build pipeline required.

• Requires Node.js 18+ and npm for building from source
• Files are served from supersplat/ directory, make sure to copy the contents of 'supersplat-build/dist/' there after building
• If a Gaussian Splat is not loading, the server hosting it may be down

• Built on PlayCanvas rendering engine with integrated SuperSplat for 3D Gaussian Splatting
• Uses .ply format for Gaussian splat data and GeoJSON for polygon geometry
• Hardware-accelerated polygon rendering with the PlayCanvas API
• Modular JavaScript architecture with manager-based pattern for state management and event coordination

The digital twin platform directly supports Ecodash's mission to "Cultivate thriving ecosystems across the planet through computational ecology and human creativity" by providing:

• Design Intelligence: Tools that translate ecological science into actionable landscape design
• 3D Ecological Digital Twins: Photorealistic simulations showing landscapes across seasons

Active development continues to enhance the ecological digital twin platform for landscape professionals. Key priorities include:

• Supply Chain Integration: Direct connectivity with native plant nurseries, enabling designers to specify plants based on ecological suitability and verify real-time availability
• Design Annotation System: Interactive markup tools for polygon overlays, allowing landscape architects to document design rationale and visualize implementation phases
• Enhanced Photorealism: Higher resolution Gaussian splat reconstruction pipelines for improved visual fidelity and design precision

Copyright Ecological Intelligence, Inc.

• SuperSplat: Advanced 3D Gaussian Splat rendering and visualization
• Native Plant Community: Growers, researchers, and designers preserving genetic heritage
• Landscape Professionals: Practitioners shaping millions of acres annually for ecological function