PatternPortrait is an innovative art installation that combines AI technology with physical artwork creation. Visitors to the installation have the unique opportunity to see their own pictures transformed into stylized line drawings, which are then painted by a robot.
The art installation utilizes a graph neural network, which has been trained on a database of bleeptrack's pattern drawings. This neural network is able to replicate and expand upon these patterns, resulting in the creation of the final drawings.
The project is still a work in progress and will continue to evolve over time. It is based on the research conducted by bleeptrack as PhD student specializing in Machine Learning and Computational Creativity. The research and so also this art piece focus on the artist's empowerment: the AI is only trained on one pattern drawing and also conserves the artist's unique drawing style. This stands in vast contrast to the current advances in AI image generation, where enormus and intransparent datasets of images are used as training material.
PatternPortrait represents the intersection of art and technology, showcasing the potential of AI in the creative process. It is a testament to the ongoing advancements in Machine Learning and Computational Creativity, and serves as a platform for exploring new possibilities in the world of art.