It is said that “beauty is in the eye of the beholder” – but so is authenticity. With the advances of generative Artificial Intelligence (genAI), art is not only being produced by the “human hand” but also through software and hardware. When we blend genAI tools – such as Chat-GPT – with a robot that can paint, the result is a system that can create art that resembles human-produced. It thus becomes crucial to identify and measure the authenticity of the artwork production process.
A painting is considered authentic when it is executed in the style, with the materials, and by the production process that is essential to the artist or genre. Despite variations in artworks, we can still identify if a painting is from the same artist by studying the style and the materials of the final painting. However, the authenticity of the production process of the artwork is still yet to be discovered – which is what we propose to do in this novel (and authentic!) project.
In the first stage of this project we plan to:
-
- Reproduce and test the FRIDA robot system currently in open access
- Explore what bio-metric sensors (e.g., motion capture, heart beat, brush stroke) can model the human and robot painting process
- Conduct interview studies with an artist in residence to capture early feedback on the robot performance and interaction design between the artist and the robot
Future stages of this project will include:
-
- Integrate genAI tools and bio-metrics with the FRIDA system
- Perform formative design studies of the FRIDA system with an experienced artist whose style has been developed
- Test if authenticity can be modeled with the data collected and perform interview studies with domain experts and artists
The UARTS Faculty Engineering/Arts Student Team (FEAST) will be engaged in achieving the following goals:
-
- Set-up the robot painting system
- Explore dynamic bio-metrics that an artist can use to model the creative process of painting
- Working with Large Language Models (LLMs) and genAI
- Perform user studies with artists
- Document and disseminate the work to the broader public and to academic audiences
Impact
-
- Exhibition on campus, publication of the work in journals/conferences
- Documentation of project and submission to conferences, journals, and festivals
- National Exhibition in museums/galleries, talks from research and artists’ perspectives, paper presentations
Students apply to a specific role on team as follows:
Biometrics Specialist (2 Student)
Preferred Skills: Experience with building analog and digital circuits, especially using sensors to capture data using Arduino/Raspberry Pi. Embedded system programming and soldering are also required.
Prerequisites: EECS 215 or equivalent
Likely Majors/Minors: CE, CS, EE, ECE, ME, ROB
Robotics Specialist (1 Student)
Preferred Skills: Experience with programming robotics systems and a basic understanding of controls. Experience in programming languages like Python or C/C++.
Prerequisites: ROB 311 or equivalent
Likely Majors/Minors: CE, CS, EE, ECE, ME, ROB
Integration Specialist (1 Student)
Preferred Skills: Proficiency in a programming language like Python, Java, or JavaScript along with understanding RESTful APIs and web development basics. Experience with using Github.
Prerequisites: EECS 183 or equivalent
Likely Majors/Minors: CE, CS, DATA, EE, ECE, ROB, SI
AV Editing & Production (1 Student)
Preferred Skills: Experience with video/sound recording and editing, use of tools like Audacity, Adobe Audition, Premiere Pro, etc. The successful candidate will work on the team to document results to create a presentation for dissemination.
Likely Majors/Minors: ARTDES, CE, CS, EE, ECE, FTVM, PAT
Faculty Project Leads
Patrícia Alves-Oliveira is an Assistant Professor in Robotics at the University of Michigan. Previously, she was a Senior UX Designer for the Astro robot at Amazon Lab126. Patricia designs interactions for social robots that empower and enhance human well being. Her interdisciplinary background unifies the fields of robotics, design research, and psychology. Patricia was a Postdoctoral Research Associate at the Computer Science and Engineering Department at the University of Washington. She received her Ph.D. from ISCTE-University Institute of Lisbon and spent time at Cornell University as a Visiting Graduate Scholar. Her research received two Best Paper Awards at the International Conference on Human-Robot Interaction. She co-founding Talking Robotics and is now a volunteer in Open Style Lab.
Abhishek Narula is an artist, designer, and engineer who currently works at the Department of Robotics at the University of Michigan. Approaching technology from a creative perspective, he investigates the aesthetic possibilities of algorithms, robotics, and interactive technologies. He earned his MFA in studio art from the University of Michigan and his MS/BS in electrical engineering from the Georgia Institute of Technology. Abhishek has exhibited at national and international venues, as the Science Gallery Detroit, Speculum Artium Media Festival Slovenia, New Media Caucus (NMC), Boulder Museum of Contemporary Art, The Boulder Public Library, The Hyde Park Art Center, and Sector 2337 Art Gallery & Printing Press, Tangible Embedded Interaction (TEI) and International Symposium of Electronic Arts (ISEA).
Students: 4
Likely Majors/Minors: ARTDES, CE, CS, DATA, EE, ECE, FTVM, ME, PAT, ROB, SI
Meeting Details: In Person. 1-hour weekly meetings in the Ford Robotics Building FRB2162. Meeting Times TBD.
Application: Consider including a link to your portfolio or other websites in the personal statement portion of your application to share work you would like considered as part of your submission.
Summer Opportunity: Summer research fellowships may be available for qualifying students.
Citizenship Requirements: This project is open to all students on campus.
IP/NDA: Students who successfully match to this project team will be required to sign an Intellectual Property (IP) Agreement prior to participation.
Course Substitutions: CoE Honors