In a project that focuses more on the process of creating art than the results, Adrienne used video screen capture to record artists drawing on the computer. Each had to start from a curved line they were presented with, and produce a drawing related to birds. Adrienne examined how different artists approached this challenge, and how they responded to other artists' attempts at it.
"Land of If" by Aubrey Millen
Aubrey created an animated short, which used flash to move drawings she had done on paper and cleaned up digitally. I had the honor of acting as the voice of the narrator for her film and behind the scenes, I was given some insight into her thought process in this piece (I also learned that my French accent is horrible). Aubrey was originally inspired by a piece of art that visually represents word frequency and distribution in a text, called TextArc (http://www.textarc.org/appearances/InfoVis02/InfoVis02_TextArc.pdf). Her idea was to create an animated storybook that followed a similar idea, using importance or frequency of different words to determine the relative size. However, it developed into a less concretely symbolic form, where concepts were visualized, not words, and were woven into the plot of the story, instead of being represented purely visually. Her story was about a little girl who, while debating whether or not to go to school, fell into an imaginary world of possibilities and choices, and encountered a potent moral dilemma. A biblical allegory, an examination of ethics, or a simple fable? Aubrey's work can be interpreted on many levels.
Andrea Mendoza's "Railroad" is a virtual reality adventure, not through an imaginary landscape, but through the history and culture of Ecuador. This interactive video installation responds to prompts by the viewer, showing archival footage shot from a railroad train when the viewer is in motion, and cutting to clips of Ecuadorian people and landscapes when the viewer stops. These clips, which are desaturated, regain color in the shape of the viewer's silhouette. And finally, the viewer can "touch" still images on the screen to enlarge them and view them more closely. It is an exploration of culture and national identity that, rather than merely lecturing the viewer on information, invites them to become part of the journey, and experience another country as if visiting.
MARCH OF THE ROBOTS by Basar Gulcu
This piece, which employs robotics and programming made from scratch, seeks to answer the question "can a robot dance?" Sure, it might not be a question that many of us were asking to begin with, but it does present an interesting quandary. The most basic and primitive form of human expression is movement. Animals use movement as a means of communication (bees even perform something like a dance), and we have adapted it into an art. To create an artificial intelligence that can mimic this behavior would be no small feat, and Basar achieved it by building a program based on genetic evolution. Basar's two robots can perform eight different movements: forward,turn right, turn left, rotate right, rotate left, wait. The sensor configuration allows the robots to sense objects (obstacles) including the partner. A pattern of movements forms one dance. Dances are evaluated as part of a Genetic Algorithm. Each agent evolves on its own where they interact with each other while learning. Hence we implemented co-evolution.
To design such a system, the factors that constitute "dance" had first to be defined. From
Basar's description of the process:
Individual Factors:
• Flow
• Directional Movement
• Adaptation
• Gathering
Performance Factors:
• Flow
• Weight
• Time
• Space
The Genetic Algorithm is a search technique, which creates a set of possible solutions to a given problem. In this case, it’s dance. The set of solutions are each a dance performance. The set of solutions are each evaluated in order to form a new set of solutions which is called a generation. In this process, they are evaluated using the dance factors. One can change the kind of dance one wants by changing the weights of the different dance factors.
Better dances have a higher probability of contributing and evolving to the next generation, because Venus and Mars learn to dance together, this can be called co-evolution.
Basar collected results from three separate generations on the robot. At Generation 0, they couldn't perform much. At Generation 100, the learning took place and they seemed to interact with each other within the given description. At Generation 300, their dance was good enough to be evaluated as of humans.
JESSICA R. EBERT
Jessica made an installation that took advantage of modern technological potential to react to stimuli from viewers, a digital image that would change at their behest. First inspired by Camille Utterback's Untitled 5: External Measures Series, in which the movement of the user is tracked by a camera and traced on a canvas. Jessica's piece, instead of using motion capture , uses speech recognition for interactive triggers. It has a photo background composited from 10 different photos she took in New Zealand, and uses animation to react to keywords spoken by the viewer. Each animation is accompanied by a sound as well. The user can affect a bird, a waterfall, and the time of day in the image in front of them. When the user says "bird," a kea, which Jessica modeled and animated using MAYA (and which, she tells me, took most of her final semester to perfect), flies into or out of the image. The kea has two potential actions, singing and laughing. The waterfall, when called out by the viewer, flows and becomes slightly transparent with each triggering. The user can also switch the image from day to night.