基于隐马尔可夫链和支持向量机人脸识别混合模型的视频节目聚类标注

Video Program Clustering Indexing Based on Face Recognition Hybrid Model of HMM and SVM

人脸在视频节目中代表了重要语义信息 ,提出使用支持向量机和隐马尔可夫链混合模型对人脸进行识别 ,然后把识别结果进行高斯聚类 ,实现视频节目的内容标注 .具体步骤如下 :首先建立人脸肤色模型 ,对视频图像中可能的人脸区域进行定位 ;从定位区域提取人脸各个器官的独立基特征 ,然后使用支持向量机和隐马尔可夫链混合模型对定位区域进行人脸识别 。

Human face is a very important semantic cue in video program. This paper presented an approach to implement face recognition with Support Vector Machine( SVM) and Hidden Markovia Model (HMM) hybrid model and clustered the outcome based on Gaussian clustering to realize the video program content indexing. The task consists of the following steps: first, the model of human face skin color is set up to locate possible human face in video frame; secondly, SVM and HMM hybrid model is used to recognize human face by independent basis feature of face apparatus; last, the recognized faces are clustered for video content indexing with Gaussian clustering.