video semantic annotation

基本解释

The semantic retrieval of news video is implemented by the annotation of scene through metadata annotation.
通过引入元数据的标注完成对场景的标注，实现用户的语义查询。
In this paper, we discuss the problem of using semi-supervised learning method to do video semantic annotation.
本文讨论了利用半监督学习方法进行视频语义标注的问题。
At present, most of video semantic annotation methods are based on statistic theory. The methods use supervised learning method to do semantic label.
目前已有的视频语义标注方法多是基于统计学理论，采用全监督学习方法进行语义标注工作。