Close
0%
0%

FestiValle 2021

A deep-learning based approach to stage lighting control

Similar projects worth following
105 views
0 followers
  • Description
  • Details
  • Files 0
  • Components 0
  • Logs 0
  • Instructions 0
  • Discussion 0
The project was developed for the "Project Course" at Politecnico of Milan in collaboration with Festivalle, a jazz festival in the Valle dei Templi in Agrigento (Sicily). We aimed at automating the process of controlling the color of a light system by mean of a neural network able to predict valence and arousal, two values used in the Russell's model of affect to describe emotions. The network is built inside a VST built in JUCE, that is fed with raw audio. It segments the stream in windows of 500ms, performs the prediction and outputs two values that are mapped to a RGB color and finally sent through OSC protocol to Madmapper. Madmapper then maps the color messages to lights through the DMX protocol.

Enjoy this project?

Share

Discussions

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates