What is more important? the technological device, the person or the selfie? This work raises the question when facing the audience to take a selfie as part of an artistic installation, -it is not just any digital self-portrait-, it is the creation of a new visual context – born from the research archives of the different departments of the EPN (Escuela Politécnica Nacional del Ecuador) – in an analog-digital-interactive piece.
Through the use of a deep convitational network and the training of the algorithm with photographs of thousands of people, the program will recognize the person and isolate them from its background; creating a new photograph, a real-time technological collage with the ability to be shared instantly through Instagram @epnselfie
This idea stems from my approach to the selfie as an object of self-significance that goes beyond interactivity with the human being or what he/she can do or understand about it.
The technical challenge of this project was to identify and classify an image in real time (semantic segmentation). The model used is based on the methodology of L. Chen, G Papandreou (2016). This model builds a maximum-hope (EM) estimator from a semi-supervised learning process. It makes use of a deep convolutional network and a set of training data with hand-labeled images.
*This art piece was commissioned for the “Reprogramar la(s) Materia(s), an Art Show part of the celebrations for the 150 years of creation of the EPN, October, 2019.
“Object-selfie-machine” would not have been possible without the support, collaboration and hard work of the following people:
David Villacis, Doctoral Student in Applied Mathematics
Laboratories that provided the images:
Acoustics Laboratory – Pamela Rivera
Meb Demex – Alicia del Carmen Guevara
Paleontology – José Luis Román
With special thanks to:
Ana María Garzón Mantilla – Curator
Juan Carlos de los Reyes
María Fernanda Orquera