Academic
Publications
Learning Functional Object-Categories from a Relational Spatio-Temporal Representation

Learning Functional Object-Categories from a Relational Spatio-Temporal Representation,10.3233/978-1-58603-891-5-606,Muralikrishna Sridhar,Anthony G.

Learning Functional Object-Categories from a Relational Spatio-Temporal Representation   (Citations: 6)
BibTex | RIS | RefWorks Download
We propose a framework that learns functional object- categories from spatio-temporal data sets such as those abstracted from video. The data is represented as one activity graph that encodes qualitative spatio-temporal patterns of interaction betw een objects. Event classes are induced by statistical generalization, t he instances of which encode similar patterns of spatio-temporal relati onships be- tween objects. Equivalence classes of objects are discovered on the basis of their similar role in multiple event instantiation s. Objects are represented in a multidimensional space that captures their role in all the events. Unsupervised learning in this space results in f unctional object-categories. Experiments in the domain of food preparation suggest that our techniques represent a significant step in u nsuper- vised learning of functional object categories from spatio -temporal patterns of object interaction.
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
Sort by: