Video Semantic Concept Detection Based on Conceptual Correlation and Boosting

Abstract

Semantic concept detection is a key technique to video semantic indexing. Traditional approaches did not take account of conceptual correlation adequately. A new approach based on conceptual correlation and boosting is proposed in this paper, including three steps: the context based conceptual fusion models using correlative concepts selection are built at first, then a boosting process based on inter-concept correlation is implemented, finally multi-models generated in boosting are fusioned. The experimental results on Trecvid2005 dataset show that the proposed method achieves more remarkable and consistent improvement.