An imaging system and method for classifying a concept type in video

AN IMAGING SYSTEM AND METHOD FOR CLASSIFYING A CONCEPT TYPE IN A method and associated imaging system for classifying at least one concept type in a video segment is disclosed. The method associates (420) an object concept type in the video segment with a spatio-temporal segment of the video segment. The method then associates (450) a plurality of action concept types with the spatio-temporal segment, where each action concept type of the plurality of action concept types is associated with a subset of the spatio-temporal segment associated with the object concept type. The method then classifies the action concept types and the object concept types associated with the video segment using a conditional Markov random field (CRF) model where the CRF model is structured with the plurality of action concept types being independent and indirectly linked via a global concept type assigned to the video segment, and the object concept type is linked to the global concept type. 1R27414 1 (P144qn11 405 Tracking Fo ah410 4 Ta -For each track determine the spatio temporal segment (S, ) associated with that track 406 segment Associate a random variable (0) (S) with each spatio-temporal segment (s,) representing a type class of the object in the segment + 430 For each spatio-temporal segment S, ( ) calculate an object feature representation (xo,) Temporally segment the spatio temporal segment (s,) into a plurality of sub-segments (S,) 450 Associate a random variable (A,,) representing a type class of action with each sub-segment (SJe) For each sub-segment (Sit) calculate an action feature representation (xA,) Associate a global random variable (E) with the video segment (S) Calculate a global scene 1407 *classification feature representation (S) for (XE) Model Build a probabilistic graphical model 490 parameters using w and xE I 0i I XAi,, for all / }W and t Determine object type(s) , action type(s)49 Semantic and the scene type classification for theF tags 7-d segment ( S) by finding the MAP solution ( End )Fig. 4