fbarretto / cibermarinum

CiberMarinum

by Suzete Venturelli, Artur Cabral and Barretto

This interactive installation is based on a Reaction Diffusion GLSL shader, running in WEBGL, that reacts to a live video stream that is processed by the ML5.js bodyPix AI model. The parameters of the RD are controlled by the data provided by a set of sensors installed on an aquarium with live plants and by tide high, wind speed and ocean turbulence forecast data provided by CPTEC API.

Table of Contents

• Installation
• Usage
• License

Installation

Clone the repository into your web container

Access ciberMarinum folder home

Usage

Mapping and parameters can be adjusted to fit different setups, for example, it is possible to add new sensors to the arduino and map them to the RD parameters. RD Parameters are defined at rDef variable as follows:

/**
 * Reaction-diffusion definition object.
 * @typedef {Object} RdDef
 * @property {number} da - Diffusion rate of chemical A (fixed).
 * @property {number} db - Diffusion rate of chemical B.
 * @property {number} feed - Feed rate of chemical A.
 * @property {number} kill - Kill rate of chemical B.
 * @property {number} dt - Time step.
 * @property {number} iter - Number of iterations.
 */

Set debug=true if you want to print the parameters at the JS console.

License

This project is licensed under the MIT License.