Saving Aesthetics 2.0

Saving Aesthetics merges Azerbaijani carpet heritage with contemporary computational processes. By studying traditional ornaments and patterns, a custom trained GAN (Generative Adversarial Network) produces new compositions that echo centuries of visual language while existing only in the digital present.

Carpets in Azerbaijani culture are more than objects. They are carriers of memory, identity, and regional storytelling. Geometric forms, floral motifs, and symbolic rhythms encode histories passed from generation to generation. Each woven pattern speaks a local dialect of a shared cultural grammar.

Today, the word “pattern” also belongs to machines, where algorithms detect structures, repetitions, and relationships across vast collections of data. This work brings these two meanings together: Traditional carpet designs are analyzed and reinterpreted through a computational process, resulting in forms that feel familiar yet have never physically existed.

The project reflects on preservation in an era of acceleration. Data can safeguard cultural memory, but it can also flatten, remix, or erase it. By placing heritage inside a generative system, the work ask show traditions survive, mutate, and remain meaningful in a rapidly shifting world. This installation invites viewers to step into a space where past and future overlap, where inherited symbols are neither frozen nor lost, but continuously reimagined.