Co-Concept-Boosting视频语义索引方法

Video Semantic Indexing Based on Co-Concept-Boosting

语义概念探测是建立视频语义索引的根本方法,传统探测方法没有充分考虑语义概念间存在的复杂关系.本文充分利用概念间关系提出了co-concept-boosting方法,该方法分为三个层次:第一层是基于上下文关系的探测模型的构建,第二层是结合概念间关系的boosting处理,第三层是对boosting过程中产生的多个探测模型的融合.利用Trecvid2005数据的实验分析证明,该方法具有良好的性能以及稳定性.

Semantic concept detection is a key technique to video semantic indexing.Traditional approaches did not take account of inter-concept correlation adequately.A new co-concept-boosting approach is proposed in this paper,including three steps： the context based conceptual fusion models 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 Trecvid 2005 dataset show that the proposed method achieves more remarkable and consistent improvement.