Tiquicia Gráfica is a surreal exploration of the ever changing graphical identity of Costa Rica. 

Since Costa Rica is a very young country, defining it’s identity and exploring it through the diverse lenses of design, is always an interesting topic.

For this project we were inspired by Costa Rican popular hand-painted signs. The main goal was to explore new ways to create art forms, through the usage of Machine Learning and GANs as tools to explore the graphic language that surrounds us.

Regarding the database, the team worked in collaboration with IdenticaCR: an archive of handmade Costa Rican signs and posters. More than 7000 images from IdenticaCR’s Instagram profile were downloaded to become the core database for the project.

The images were sorted in four main categories: faces, characters, signs and food. After that, four different GAN models were trained with each category, allowing the team to explore in detail the patterns and aspects of these graphic expressions.

The outcome was an interactive exhibition that involved both a gallery with more than 100 print outs and an interactive projection mapping that allowed users to blend in with the new graphical concept. Metaphorically speaking, the visitors could merge their bodies into this new graphic language generated through the GAN images.

Machine Learning was a tool to spark up creativity that allowed the team to explore exponentially one small part of the Costa Rican graphic identity.

The intent behind it

Costa Rica is one of the most biodiverse countries in the world but without a clear understanding of its own visual identity. The intent was not to define our graphic identity – and we can’t–  but it is our responsibility as designers to keep on exploring and questioning it: Who might feel represented with these new graphic expressions generated by the machine? Is this part of our identity?