Multimedia Event Detection Using Hidden Conditional Random Fields

This paper introduces a method for Multimedia Event De-tection (MED). Given training videos for a certain event, a classifier is constructed to identify videos displaying it. In particular, the problems of the weakly supervised setting and the unclear event structure are addressed in this paper. The first issue is associated with the loosely annotated training videos that usually contain many irrelevant shots. The sec-ond one is the difficulty of assuming the event structure in advance, because videos can be created by arbitrary cam-era and editing techniques. To overcome these problems, a Hidden Conditional Random Field (HCRF) is used where hidden states work as an intermediate layer to discriminate between relevant and irrelevant shots to the event. In ad-dition, the relation among hidden states characterises the event structure. Thus, the above problems are managed by optimising hidden states and their relation, so as to distin-guish videos where the event occurs from the rest of videos. Experimental results on TRECVID video data validate the effectiveness of HCRFs in this context