Our installation explores interesting relationships of those in the space using the identity and physical presence of each individual as the catalyst for events in the installation. More abstractly, we address ideas in personal identity and perception of identity.
The site consists of a single space that is subdivided into four smaller areas: the entrance, main video wall, individual room, and mating room. Cameras are placed throughout the space to achieve total coverage with many viewing angles. Also, speakers are placed throughout all points of the space. The figure here shows how the space is to be laid out:
While in the entrance and before entering the main part of the installation, visitors must provide 3 items describing their identity:
These items are "associated" (i.e. by a computer controlling the installation) with a location tracking tag each visitor is given to identify themselves as they wander through the installation.
Several large projection screens, a set of speakers, and many video cameras interact to dynamically display images and sounds based on the information taken from the scans, surveys, and recordings garnered from each person who enters the installation. Each of the pieces described in this section are also doubled and thus can be seen in real time on the installation's web site.
This composite is shown on the large video screen that is the first and dominant thing seen right after visitors pass through the entrance. The real-time video canvas will take video feeds of each person from the many video cameras and algorithmically superimposes them like so many pieces of animated tracing paper. Although the resulting image might become unrecognizable as the number of viewers scales up, it will instead become an entirely different type of aesthetic display. This composite can explore the issue of crowds with the suspension of reality where many people apparently exist in the exact same physical position.
The still passport-like snapshots of each viewer are dynamically morphed into a still portrait of a person who is the fusion of the appearances of all people within the installation. Issues such as the meaning of individuality, perceptions of androgyny, and relationships to others are addressed.
This image is simply for illustration purposes and is taken from the CG facial expression work done by Frederic Pighin, Jamie Hecker, Dani Lischinski, Richard Szeliski, and David Salesin. The reference is:
Frederic Pighin, Jamie Hecker, Dani Lischinski, Richard Szeliski, and David Salesin. Synthesizing Realistic Facial Expressions from Photographs. Proceedings of SIGGRAPH 98, in Computer Graphics Proceedings, Annual Conference Series, 1998.
Using a touchpad interface and a survey of general characteristics, viewers automatically construct a police sketch of themselves from a catalog of common features. The goal is to explore formulaic physical description such as "I am tall" or "I have green eyes". The advantages and inadequacies of such abstract generalizations are illustrated by showing sketches of a figure that should be you by symbolic description yet probably does not resemble you.
Note: This sketch is from the FBI (www.fbi.gov) wanted list.
Two side-by-side projection screens display life-size images of a male and a female figure. These displays are photorealistic instantiations of the average ideal mate surveys provided by people currently within the installation. This display should raise interesting issues in defining physical beauty on a group versus personal level and may address certain stereotypes about how we view each other. A possible extension is to have another display showing the result of averaging all people who have ever been in the installation instead of just the current folks.
With this component we simulate "mating by proximity." A baby bassinette with a video screen mounted inside displays a composite image of what resulting child might look like if the people with the closest proximity to the crib were to mate. To see the image, one must approach and peer into the crib. The awkward moment that results when the viewer sees their influence on the image relative to someone already standing there is the desired effect and brings up interesting issues of individuality, personal identity, and sexuality.
In a similar way that we have mixed and synthesized visual characteristics of the visitors, here we synthesize their vocal characteristics. Each visitor digitally records their name and a single note at the entrance. A special vocoder analyzes the source and distributes the vowels (and some consonants) into eight pitches, according to the predetermined scheme (see the musical example below).
The German composer Karlheinz Stockhausen (b. 1928) based his "Stimmung" (1968) work for 6 vocalists on the B-flat ninth chord: Bb2, D3, F3, Ab3, and C4. In order to provide the visitors with something ambient and static, we use a similar, but slightly more complicated, chord structure, the B-flat fifteenth chord: Bb2, D3, F3, Ab3, C4, Eb4, Gb4, and Bb4. You can hear the example here [chord.mp3].
Each vowel is uttered with the consonant(s) preceding it (if any) on the predetermined pitch when the synthesis is ready (the calculation time is approximately 0.00001 second, so this is virtually a real-time synthesis), and played continuously until the visitor whose name is used for that vowel exits the installation. Then the consonant(s) following the vowel is finally uttered. Let us give you an example: "Hideaki Onishi" would give you a D3, an Ab3, a C4, and four Bb4s, and would sound like this (the consonants are not included in this example) [hideaki.mp3].
When there are more than one visitor and their names have some vowels in common, their voices are synthesized and result in something in-between in terms of timber. This synthesis works no matter how many visitors are present. The more visitors there are, the thicker the chord becomes and the harder it becomes for them to recognize their own voices. However, his/her voice is still a part of this ambient music as far as he/she is there, and contributes to the whole of an art work made out of their identities. Here is one more example: the mixture of our names (Hideaki Onishi, Jon Howe, David Schneider, and Jeffrey Hightower) creates the following sound continuum [allofus.mp3]. Not very exciting? Then you should visit our installation and feed your name to this system!
The whole chord structure slides up and down extremely slowly (the intervallic relationship remaining the same all the time), creating an unconscious image of a floating space of "zero gravity." The eight-(or more)-channel audio equipment surrounding the visitors also enhances this effect; the sound image moves around the installation slowly, and affects the visitors' sense of direction.
We found that the following qualities contribute to an effective multidisciplinary team: flexibility, creativity, interest in others perspectives and good follow through. The flexibility allowed us to settle on an idea which was interesting to everyone. Creativity manifests itself in a number of different ways. Specifically, we'd like to bring attention to the ability to continue to develop ideas in a group setting. One person may start an idea, and then another one develops it, and then the first person bounces off this new development. In contrast, a meeting where little is said but much was thought would be less effective. In a team where the members come from different backgrounds, being not just open to, but also interested in the different perspectives of the team is a valuable property. This helps to create an environment where the individuals start to feel comfortable and excited about contributing to the project. Finally, good follow through is clearly important for turning the ideas into a concrete project. Proficiency with tools for document creation and web page design is very helpful for this last stage.
March 10, 2000 Proposal Presentation
Seminar in Digital and Electronic Arts Homepage