Advances in Multimedia Information Processing – PCM 2012

Video-based face recognition is a fundamental topic in image and video analysis, and presents various challenges and opportunities. In this paper, we introduce an incremental learning approach to video-based face recognition, which efficiently exploits the spatiotemporal information in videos. Face image sequences are incrementally clustered based on their descriptors. With the quantization of the facial words, representation of the face image is generated by concatenating the histograms from regions. In the online recognition, a temporal matrix and a voting algorithm are employed to judge a face video’s identity. The proposed method achieves a 100% recognition rate performed on the Honda/UCSD database, and gives near realtime feedback. Experimental results demonstrate the effectiveness and flexibility of our proposed method.