Unsupervised Activity Discovery and Characterization From Event-Streams
Rafay Hammid, Siddhartha Maddi, Amos Johnson, Aaron Bobick, Irfan Essa, Charles Isbell
We present a framework to discover and characterize different classes of everyday activities from event-streams. We begin by representing activities as bags of event n-grams. This allows us to analyze the global structural information of activities, using their local event statistics. We demonstrate how maximal cliques in an undirected edge-weighted graph of activities, can be used for activity-class discovery in an unsupervised manner. We show how modeling an activity as a variable length Markov process, can be used to discover recurrent event-motifs to characterize the discovered activity-classes. We present results over extensive data-sets, collected from multiple active environments, to show the competence and generalizability of our proposed framework.
PDF Link: /papers/05/p251-hammid.pdf
AUTHOR = "Rafay Hammid
and Siddhartha Maddi and Amos Johnson and Aaron Bobick and Irfan Essa and Charles Isbell",
TITLE = "Unsupervised Activity Discovery and Characterization From Event-Streams",
BOOKTITLE = "Proceedings of the Twenty-First Conference Annual Conference on Uncertainty in Artificial Intelligence (UAI-05)",
PUBLISHER = "AUAI Press",
ADDRESS = "Arlington, Virginia",
YEAR = "2005",
PAGES = "251--258"