Siggraph 2005 logo
Conference Exhibition Call For Participation Presenters Media Registration

emerging technologies

conscious=camera



The reality we perceive is not a flow of flat images as a digital camera or video camera would output, but consciousness "of things." This project attempts to represent that consciousness.

If the movement is slow, only the face and hands are shown: this is the consciousness of a static person. If the movement becomes more important, the face and hands disappear and the body begins to appear: this is the consciousness of a less distinguishable body in motion. A trailing effect makes each image integrate the "before" and "after" moments, which are components of the consciousness of movement. Each consciousness shown leaves a mark on the background, which will very slowly vanish. This represents the memory.

Art and Science


Initial reactions to conscious=camera are playful: users want to understand what it sees and get an idea of how it works. First, users see their hands filtered, then the effect of moving, and finally the memory effect (an image of their action on the camera’s background). Thus, they understand how previous users have generated other marks, making their actions present in the mind of the current users. The conscious=camera then creates a presence, and this presence is the presence of the others. It is an ephemeral mirror that reveals parts of what we are and how we act, and makes us see with new eyes the others who surround us.

Vision


Computer vision and graphics have reached a state where they can represent, detect, and generate through images a colossal part of what is meaningful to us, at continually increasing speeds. Although many algorithms are complex, understanding their complexity gives the programmer the ability to finely manipulate their output. It is through understanding how the lowest levels of the program's elements affect and transform the aesthetic aspect that we have been able to construct an engaging result.

Goals


By oscillating between movement/motion blur and static moments, only some poses of faces and bodies appear on the screen: perhaps a profile, or a face looking toward a particular direction. These are those "privileged moments" (Bergson) we would remember from an interaction with somebody, or those "decisive instants" (Cartier-Bresson) that a photographer would like to capture. If the conscious=camera does not represent the actual biological state of consciousness, it gives at least a poetic and meaningful way of looking at how we act and how we appear. Its memory also becomes a door to interactions in the social world.

Innovations


This project relies on implementation of body-tracking techniques inspired by the MIT Pfinder. A first challenge was making the algorithm reliably track in real time the bodies of the variety of people who will experience the installation, even in extreme conditions (for instance, if the user gets very close to the camera). Another is the location of the creative center of our work: at the low-level manipulation of all the program's elements - the "inside" of the algorithm - inserted in a complex structure of filters and functions tuned to make the program produce the aesthetically desired result.

Contact


Kevin Quennesson
Georgia Institute of Technology
kevin.quennesson (at) polytechnique.org
acm.org siggraph.org