Evolution of Mona Lisa with Pablo Picasso's paintings

Abstract

In recent years, design of artistic artifacts using machine learning (ML) techniques has become a major feat. Artificial creation of artistic artifacts involves the use of ML. Employment of ML has many benefits for artificial art. On one hand, it reduces the need for human dexterity required particularly for this niche. On the other hand, artificially created art can often lead to spectacular outcomes that are normally not possible by sole human effort. In this paper, we demonstrate our work about the recreation of Mona Lisa using an evolutionary algorithm (EA). The EA uses an image corpus of Pablo Picasso's work. To this end, the images form the design space for the problem at hand. By using these images the EA attempts to recreate Mona Lisa. The reported results show the miraculous and awe-inspiring accuracy of EAs.