ABSTRACT
Motion detection is closely coupled with higher level inference tasks such as detection, localization, tracking, and classification of moving objects, and is often considered to be a preprocessing step. Its importance can be gauged by the large number of algorithms that have been developed to-date and the even larger number of articles that have been published on this topic. A quick search for ”motion detection” on IEEE Xplore c© returns over 20,000 papers. This shows that motion detection is a fundamental topic for a wide range of video analytic applications. It also shows that the number of motion detection methods proposed so far is impressively large.
