Unfolding Confusion

unfolding confusion, closeup

Unfolding Confusion

Folded Photographies, video tapes from personal archives

Valencia 2023

Unfolding Confusion explores the relationship between twins, art and artificial intelligence through a mixture of autobiographical, experimental and speculative perspectives.
Due to their strong resemblance, twins are particularly suitable for testing AI systems for their accuracy, blurriness and blanks, as they question schema formation, i.e. recognition by recognition, as an inherent component of perception.

The non-linear narrative style of the film creates associative connections between the different materials. These range from VHS recordings from the personal archive, machine vision models with personalized training data to experiments with manipulated photographs. The film always explores the question of how the perception of the human eye and the architecture of AIs for image recognition differ.

My full Master Thesis can be found in german & spanish under the given links.

 

Before and after results of prompt output via clipInterrogator by Language model of Stable Diffusion

By folding, the ViLT-Language Model (the Machine Vision Model of Stable Diffusion) recognizes only one person

Short film based on research during my master in Visual Arts & Multimedia, UPV Valencia

Unfolding Confusion explores the relationship between twins, art and artificial intelligence. through a mixture of autobiographical, experimental and speculative perspectives.
Due to their strong resemblance, twins are particularly suitable for testing AI systems for their accuracy, blurriness and blanks. They question schema formation, i.e. recognition by recognition, as an inherent component of perception.

The non-linear narrative style of the film creates associative connections between the different materials. These range from VHS recordings from the personal archive, machine vision models with personalized training data to experiments with manipulated photographs. The film always explores the question of how the perception of the human eye and the architecture of AIs for image recognition differ.