NEAR Creative Engine - Fractal Studio

This repository contains the NEAR Foundation grant implementation for real-time fractal generation with emotional computing.

Project Overview

We propose developing a Rust-to-WASM compilation module that enables direct deployment of our NUWE (Neuro-Unified Wave Environment) fractal shader engine as a NEAR Component (dApp) in the browser. This will create user-owned, real-time creative sessions that run entirely on-chain, democratizing access to high-performance creative tools as public goods.

Features

• Real-time Fractal Generation: Interactive Mandelbrot, Julia, and Burning Ship fractals
• Emotional Computing Integration: Visual parameters modulated by emotional state
• NEAR Blockchain Integration: Sessions saved and owned on NEAR
• Browser-Native Performance: WASM-compiled engine for desktop-quality performance
• Practical Emotional Input: Multiple input methods including camera-based facial expression analysis

Getting Started

Prerequisites

• Rust and Cargo
• Node.js and npm
• NEAR CLI
• wasm-pack

Installation

# Install CLI tools
./scripts/install-cli-tools.sh

# Build the project
./build-near-grant.sh

Building

# Build NEAR smart contracts
cd src/near-wasm
cargo build --target wasm32-unknown-unknown --release

# Build WASM for browser
wasm-pack build --target web --out-dir ../../test-website/wasm

Deployment

1. Deploy contracts to NEAR testnet
2. Update contract IDs in test-website configuration
3. Serve test-website on a web server

Practical Emotional Input Methods

Our system implements multiple practical approaches for emotional input that work without specialized hardware:

1. Manual Slider Controls (Primary Method)

   • Valence (Negative ↔ Positive) slider
   • Arousal (Calm ↔ Excited) slider
   • Dominance (Submissive ↔ Dominant) slider
   • Works on any device with a browser

2. Camera-Based Facial Expression Analysis (Enhancement)

   • MediaPipe FaceMesh for facial landmark detection
   • Real-time mapping of facial expressions to emotional states
   • Works with any standard webcam (built into most devices)
   • Local processing - no data leaves the user's device

3. Keyboard/Mouse Interaction Tracking (Passive Input)

   • Typing pattern analysis for arousal inference
   • Mouse movement dynamics for emotional state detection
   • Subtle emotional updates based on interaction behavior

4. Voice Tone Analysis (Future Enhancement)

   • Audio input processing for emotional inference
   • Works with standard microphones
   • Local processing for privacy

5. EEG/BMI Integration (Future Research Enhancement)

   • Compatible with consumer EEG devices (OpenBCI, Muse, etc.)
   • Advanced biometric emotional state detection
   • Research-grade precision for specialized applications

Directory Structure

├── src/
│   ├── near-wasm/          # NEAR smart contracts and WASM engine
│   └── rust-client/        # Core Rust library (shared dependency)
├── test-website/           # Browser-based frontend
├── scripts/                # Utility scripts
├── build-near-grant.sh     # Build script
└── README.md              # This file

License

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

Contact

• Website: https://compiling-org.netlify.app
• GitHub: https://github.com/compiling-org
• Email: kapil.bambardekar@gmail.com, vdmo@gmail.com