Using Motion Expressiveness and Human Pose Estimation for Collaborative Surveillance Art

Abstract

Surveillance art is a contemporary art practice that deals with the notion of human expressiveness in public spaces and how monitoring data can be transformed into more poetic forms, unleashing their creative potential. Surveillance, in a sociopolitical context, is a participatory activity that has changed radically in recent years and could be argued to produce, not only social control but also to contribute to the formation of a collective image of feelings and affects expressed in modern societies. The paper explores a multidisciplinary approach based on tracking human motion from surveillance cameras on New York Time Square. The performed human trajectories were tracked with a real-time machine vision framework and the outcomes were used to feed a generative design algorithm in order to transform the data into emotionally expressive 3D visualizations. Finally, a study was conducted to assess the expressive power of this approach so as to better understand the relationships among perceived affective qualities and human behaviors.