Unsupervised learning of action primitives
2010 (English)In: 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010, IEEE , 2010, 554-559 p.Conference paper (Refereed)
Action representation is a key issue in imitation learning for humanoids. With the recent finding of mirror neurons there has been a growing interest in expressing actions as a combination meaningful subparts called primitives. Primitives could be thought of as an alphabet for the human actions. In this paper we observe that human actions and objects can be seen as being intertwined: we can interpret actions from the way the body parts are moving, but as well from how their effect on the involved object. While human movements can look vastly different even under minor changes in location, orientation and scale, the use of the object can provide a strong invariant for the detection of motion primitives. In this paper we propose an unsupervised learning approach for action primitives that makes use of the human movements as well as the object state changes. We group actions according to the changes they make to the object state space. Movements that produce the same state change in the object state space are classified to be instances of the same action primitive. This allows us to define action primitives as sets of movements where the movements of each primitive are connected through the object state change they induce.
Place, publisher, year, edition, pages
IEEE , 2010. 554-559 p.
Body parts, Group actions, Human actions, Human movements, Imitation learning, Key issues, Mirror neurons, Motion primitives, Object-state space, Anthropomorphic robots, Unsupervised learning
Computer Vision and Robotics (Autonomous Systems)
IdentifiersURN: urn:nbn:se:kth:diva-150045DOI: 10.1109/ICHR.2010.5686309ScopusID: 2-s2.0-79851504957ISBN: 978-142448688-5OAI: oai:DiVA.org:kth-150045DiVA: diva2:742391
2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010, 6 December 2010 through 8 December 2010, Nashville, TN, United States
QC 201409012014-09-012014-08-292014-09-01Bibliographically approved