Abstract
|
TV commercials are interesting in many respects: advertisers and
psycho logists are interested in their influence on human purchasing
habits, while pare nts might be interest in shielding their children
from their influence. In this paper, two methods for detecting and
extracting commercials in digital videos ar e described. The first
method is based on statistics of measurable features and enables the
detection of commercial blocks within TV broadcasts. The second meth
od performs detection and recognition of known commercials with high
accuracy. F inally, we show how both approaches can be combined into a
self-learning system. Our experimental results underline the
practicality of the methods.
|
