Online Shot Boundary Detection Based on Incremental Principal Component Analysis

Shot boundary detection is the key point of content-based video retrieval. In order to implement online detection, a method based on incremental principal component analysis (IPCA) is proposed to analyze video in a real-time way in this paper, and then to get the shot boundary. Its advantage is detecting shot boundary and extracting Abstract video without acquiring the whole video in advance, in virtue of which it makes the online detection come true. But it needs large memory, which can be solved by the way proposed. The experiment proves it has good performance.