Relational Graph Mining for Learning Events from Video

In this work, we represent complex video activities as one large activity graph and propose a constraint based graph mining technique to discover a partonomy of classes of subgraphs corresponding to event classes. Events are defined as subgraphs of the activity graph that represent what we regard as interesting interactions, that is, where all objects are actively engaged and are characterized by frequent occurrences in the activity graph. Subgraphs with these two properties are mined using a level-wise algorithm, and then partitioned into equivalence classes which we regard as event classes. Moreover, a taxonomy of these event classes naturally emerges from the level-wise mining procedure. Experimental results in an aircraft turnaround apron scenario show that the proposed technique has considerable potential for characterizing and mining events from video.