Robot Artwork

From ISLAB/CAISR
Revision as of 19:17, 14 October 2018 by Slawek (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Title Robot Artwork
Summary Capability for a robot to paint to express human feelings
Keywords Robots, Artwork, Emotion, BMI, Pattern recognition, Computer vision
TimeFrame 2017/1/1-2017/8/30
References -robot artwork

Michael Raschke, Katja Mombaur, Alexander Schubert. An optimisation-based robot platform for the generation of action paintings. Int. J. Arts and Technology, Vol. 4, No. 2, 2011 181


-emotion recognition from eeg

Yuan-Pin Lin, Chi-Hong Wang, Tzyy-Ping Jung, Tien-Lin Wu, Shyh-Kang Jeng, Jeng-Ren Duann, and Jyh-Horng Chen. EEG-Based Emotion Recognition in Music Listening. IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 57, NO. 7, JULY 2010

Prerequisites Interest in robots and humans (artwork and emotions);

Willingness to work to help others, also by competing in a contest with a chance to earn prestige and money for charities and the university; Software skills

Author Daniel Westerlund, Sowmya Narasimman
Supervisor Martin Cooney, Maria Luiza Recena Menezes
Level Master
Status Finished


There is something profound about art. Part of it may be the freedom to visit new "spaces" and feel in a new way. Another part may be the way it captures aspects of our personalities, our priorities, our feelings, our desires toward others, our motives for doing art, how we wish to be treated, and our understanding of ourselves and the things around us--in a visceral way which can be communicated to others. (For example, young children's artwork typically proceeds in stages, from scribbles which feel good, to "functional" symbols like cylinders with minimal appendages to represent humans, to "logical" renderings in which all body parts are drawn, to "realistic" shaded renderings; likewise, differences have been found in adults in drawings of nerve cells by students and experts.) In this project, the main idea is to have a robot create a painting based on something recognized from a human. There are many possibilities to do something useful in this context. For example, art therapy can help people to feel better. Using sensors, the feelings of people with autism can be conveyed, who may have problems communicating feelings socially. Using a robot Parkinson's disease patients could draw straight lines without tremors. Feedback can be incorporated while painting to make changes; a caricature could be made more or less exaggerated, or a sketch more or less abstract, based on if a human looks happy or unhappy with progress. Neural networks could be an interesting mechanism to express emotions, like Google's work with "inceptionism"...


Some notes:

  • One goal of this project is to produce a submission to the RobotArt contest (http://robotart.org/, founded by Andrew Conru), which offers $100,000USD in various prizes (25% university, 75% charities in the US). The deadline for submitting art will be April 15th, with a decision in May.
  • The robot used will (probably) be the Baxter robot, an advanced humanoid robot with two seven degree-of-freedom arms; the student will receive time each week to work with the robot, but time with the robot will also be shared with other students and researchers as needed.
  • The sensor used will (probably) be a brain-machine interface.
  • We will seek to obtain advice from various experts, e.g., in art, psychology, or computer vision.
  • The research focus will be decided after initial discussion.


Preliminary workplan:

  • Learning how to use the robot and sensor.
  • Literature review.
  • Building a system.
  • Evaluating the system
  • Working on thesis and presentation

Focus on software or hardware?: software

Expected results: a thesis, code, video, submission to RobotArt Competition (it would also be nice, but not required, if the student is willing to also write a six page shortened version of the thesis, to be submitted to a conference)