Bischof, W. F. and Caelli, T. (2001). Learning spatio-temporal relational structures. Applied Artificial Intelligence, 15, 707-722.
We introduce a rule-based approach for learning and recognition of complex actions in terms of spatio-temporal attributes of primitive event sequences. During learning, spatio-temporal decision trees are generated that satisfy relational constraints of the training data. The resulting rules, in form of Horn clause descriptions, are used to classify new dynamic pattern fragments, and general heuristic rules are used to combine classification evidences of different pattern fragments.
Back to publications.