Inside-Out

Inspiration

Inside-Out is an exploration in the creative processes surrounding the Machine learning in the art world. It is a number of works that combine together into one overarching project. Some finished and some didn't. However, the continual exploration of this idea is what we really understand as the process of Art.

What it does

This project explores the complex nature of neural networks and the process around. It first used feedback looping to create a multitude of video artistic pieces that show the propagation of data through a system. It also has a live component as well. Finally, I was finishing work on creating a method to allow a computer to train a human rather then the reverse. The final is still being explored, and I hope to continue to explore it at the demo with those around me.

How I built it

The feedback loops are created with 2 web cameras in a double feedback loop method, which allows me to add addition delay into the system. The machine learning algorith was started with TensorFlow and OpenCV.

Challenges I ran into

One of the major obstacle that I ran into was the amount of time it would take to train. Normal training for Neural Networks takes lots of time, but it takes even long to do so with a human. To develop these connections quickly and efficiently without having any prior knowledge of machine learning was extremely difficult.

Also, Devpost does not allow video files for submission, so it's a little hard to post the exact artworks. Never the less, they will be up as soon as possible.

Accomplishments that I'm proud of

I'm extremely proud that I was able to showcase some of my works live at

What I learned

What's next for Inside-Out

Built With

• logictechwebcamera
• tensorflow