Prosopagnosia Project

Joan Fontcuberta / Pilar Rosado



 "We venture through a series of conflicts and violences of our time, while always trying to save what still makes us human and commits us to values that reject barbarism".

"... An alpha and an omega: from the ruins of photography – the images from family photo albums that are damaged and become amnesic, while reflecting their most terrible beauty – to the predictions of possible futures and algorithms, which we do not know whether they will be nightmares or guarantees of progress".


Prosopagnosia is an artificially created speculation on the dialectic of facial recognition. Generative Adversarial Networks (GAN) have been used, which are deep neural network architectures composed of two networks that face each other (therefore, theterm "adversary"), to create portraits that have never existed. Ian Goodfellow and other researchers from the University of Montreal introduced GANs in an article in 2014. The potential of these models is enormous because they can learn to mimic any dataset, that is, GANs can be taught to create haunting worlds, similar to ours in any domain: images, music, speech, or prose.

Joan Fontcuberta and Pilar Rosado share concerns about art and technology that fuel their collaboration. Both artists have been developing the project Prosopagnosia, related to facial recognition, that addresses the ontology of photography by generating portraits of people who don’t exist. The images are created by Generative Adversarial Networks (GAN), algorithms that have enabled us to achieve an unprecedented degree of realism in the creation of synthetic photographic portraits. The system GAN works by pitting two neural networks against each other in a competitive training. The generative network tests out combinations of elements from a latent space to generate candidate images to be similar to those in the database. The other network, the discriminator, assesses the candidates put forward by the generative network using the original data in order to decide which faces are genuine and which are artificial.


Joan Fontcuberta and Pilar Rosado

Joan Fontcuberta (Barcelona, 1955) and  Pilar Rosado (Barcelona, 1965).

Joan Fontcuberta has developed both an artistic and theoretical activity, focused on the conflicts between nature, technology, photography, and truth. He is a self-taught artist and his training is mainly in communication and social sciences. From that theoretical starting point, his work has focused on the changing nature of images. However, beyond aesthetic qualities, images are understood as social and historical constructions that provide models for the real world and allow human interaction.

Pilar Rosado is an artist, teacher, and researcher. Graduated in Biology and PhD in Fine Arts, she has published various essays on the application of computer vision models for the analysis of large collections of abstract art images, which provide alternative points of view for reflection and which question the conventions of our gaze. In her artistic practice, she explores political issues that can be addressed from the image and that implicate machine learning technologies, such as information management in the visual archives of the future, revision of collective memory, or artificial creativity.