Research And Practice Of The Technology Of Video Retrieval Based On Face Recognition

With the rapid development of multimedia and computer network technology, digital video and multimedia applications emerged at an extremely high rate in computer network. It is of great significance to meet user’s need for effectively organizing, managing and using these video data, and content-based video retrieval has been the most active research focus in current multimedia field. In this paper, we focus on the technology of video retrieval based on face recognition, and it has research and application value in multimedia management.In this paper, we reviewed the technology related with face recognition and video retrieval. First, we studied Viola-Jones face detector and a method based on local constraint feature model in face extraction process. Second, we proposed a method for front facial judgment, which had been verified to be accurate through experiments. Finally, we introduced a modified face recognition technology based PCA to get person’s information. Also, we designed and implemented a video retrieval system. According to experiment, the system we designed in this paper can achieve quite high retrieval accuracy in news video.