For thousands of years, art is the way to express our deepest feelings. ART is understood as any activity that aims to raise emotions, communicate your inner being with the outside and capture them in something beautiful. Painting, music, reading, among others, will be part of these ways of expressing our passions.Currently the world of technology is a way of expressing ourselves and why not mix with other ways to make art and create something EPICDARY, epic and legendary.

Our project consists on the understanding of the style context of an artist using neural networks, based in the way and particular characteristics of each one. Painting like Leonid Afremov, Francis Picabia or others is difficult, but with this tool you can feel that your image is created with the same way of thinking of this artist. IMPORTANT TO SAY THAT IS NOT A FILTER, IS A WAY OF UNDERSTAND OF HOW EACH ARTISTS USE THEIR TECHNIQUES IN EACH ART.

The first step was work with object detection with OPENCV to detect the art that we want convert in the artistic style.We decided to use our camera to detect the objects to make an interactive application. Then we use neutral networks to adapt each one to a special artistic technique.

One of the problem we had was lost a lot of time training the model and making a lot of test. I learned a lot of openCV, art, extend my knowledge of neutral networks.

I am completely happy to see this application running and see that my images looks like the artists's technique. It was awesome to see the results.

What's next for NeuroARt: The hardcore is done, so the only thing is missing is making a better interface app and creating new artistics styles.

Built With

Share this project: