Decisive Camera

image classifier, website, installation, 2017+2018

Full documentation coming soon

Decisive Camera is a new chapter in a series of works by Sebastian Schmieg raising questions about photography and machine learning technology. Machine Learning can be defined as a subfield of computer science and artificial intelligence, that studies the possibility to give machines the ability to learn without being explicitly programmed. It makes use of algorithms that, based on a set of data, can learn to make predictions, to classify and to make decisions.

The first part of the project, produced for The Photographers’ Gallery (TPG), was based on the gallery’s image archive available online. Visitors were invited to assign all the images to one of four given categories: Problem, Solution, Past and Future. Decisive Camera moves a step forward asking Link Cabinet visitors to take a picture with their own webcam or to upload an existing image. The system will then classify each image, based on the dataset created in the first part of the project, and it will state to which percentage each photo responds to the categories of Problem, Solution, Past and Future.

Schmieg’s work is a multifaceted experiment that questions the role and value of artificial intelligence, human intervention, the biases embedded in algorithmic systems and the ability of machine learning to confront with complex and abstract concepts.

(Text from Link Cabinet / Matteo Cremonesi)