Athletics images interpretation using structural ontology model

The continual rapid growth in digital content acquisition and visualization makes it increasingly challenging to find, organize, and access visual information. Typically, image classification and retrieval systems tend to rely only on the lowlevel visual structure within images. Image classification methods usually perform based upon a vector space model. This paper presents a framework to restructure the vector space model of visual words with respect to a structural ontology model in order to resolve visual synonym and polysemy problems. The experimental results show that our method can disambiguate visual word senses effectively and can significantly improve classification, interpretation and retrieval performance for the athletics images.