Measuring Similarity between Trajectories using Motion Verbs in Semantic Level

Nowadays, the emerging technology for video retrieval is mainly based on the semantic in the high-level. In the previous works, we proposed a novel approach for motion recognition from the aspect of semantic meaning. This issue can be addressed through a hierarchical model that explains how the human language interacts with motions. In this paper, we describe similarity measure with link type based on our proposed model. Specially, we focus on antonym relation except for basic relations (IS_A, PART_OF) in hierarchical model. And we got the most suitable weight value (0.7) through correlation coefficient. We experiment semantic based trajectory retrieval using the proposed similarity measure considering the weight value.