Dancing Around the Subject: AI Co-Creative Research Continues with KSU Motion Capture Sessions
Posted December 7, 2022
Note: This story was prepared by Kennesaw State University Communications and was first published on the KSU website.
Wriggling into a stretchy, full-body black suit with internal wires, Kennesaw State University senior Jordan Carter smiled.
“I feel like a spy,” the dance major said as she zipped up the motion capture suit, put on the accompanying gloves and strapped an elastic crown to her head.
Nearby, KSU Assistant Professor of Dance Andrea Knowlton watched two stick-figure avatars on a laptop screen. Carter and junior Tyler Hayes admired their suits and made adjustments before going through a series of predetermined movements.
The exercise was part of a research project Knowlton and her students are working on with a team of researchers from Georgia Tech, led by Georgia Tech Professor of Digital Media Brian Magerko, who reached out to Kennesaw State’s Department of Dance while applying for external funding.
Knowlton and Magerko received grants from the National Science Foundation for the project, and Knowlton is working with her students to study dance applications in teaching movement to artificial intelligence through improvisational dance. Their collaborative research gathers information from human dancers to better understand co-creativity. They are applying this knowledge to the development of AI agents who can actively participate in physical collaborations with dancers. The work will culminate in a live AI dance performance with students and faculty in KSU’s Department of Dance when the grants end in September 2024.
“The partnership is really a win-win. Brian and his team needed access to dance majors, studio space, and students committed to the research process. I have those things, and as I’m a bit of a technophile, it was exciting for me to get to collaborate on this research,” said Knowlton, who wrote her master’s thesis on interactive wearable software and sensemaking, the process by which people make sense of their environments, through dance.
According to Knowlton, the collaboration with the Georgia Tech researchers has taken her work in a new direction and has opened her eyes even further for the potential interactions between the seemingly disparate fields of science and dance.
“It’s stunning how many things we just don't understand about each other’s expertise,” Knowlton said. “We're always filling each other in. It’s really raising the level of the research.”
Through the grant, Knowlton has built her own research team and brought senior dance major Alexis Young on board as a research assistant. Young plans a career in physical therapy, using her dance experience as a model for movement and bolstered by the technology she has worked with on this study.
“For a dancer, using artificial intelligence to see movement patterns and stylistic qualities gives a new physiological and biomechanical understanding of our own bodies,” Young said. “The combination of both dance and science provides limitless discoveries, and with the advanced technology that we have today this combination can be explored in depth.”
Recently, the Georgia Tech researchers visited Kennesaw State to electronically capture the movements of Knowlton’s improvisational dance class.
With the dancers lighting up the small computer screen, the researchers measured and recorded each movement. Occasionally, the dancers would catch a glimpse of themselves as stick figures, reacting with delight at their motions boiled down to colorful avatars.
“This is research,” Hayes said. “It’s amazing to be a part of it.”
According to a Georgia Tech press release, the project, scheduled to run three years, is expected to lay the groundwork for the development of other applications that could benefit from such improvisational abilities, including physical therapy, design brainstorming, or future experiences with robots in the home.
— Dave Shelles, Kennesaw State University