Revealing the connoted visual code: a new approach to video classification

In this paper, we present a new approach for classifying video content into semantic classes at a high level of abstraction by exploiting the connoted visual code. The method is based on the concept of supervised learning algorithms that have already been applied for the classification of written text and spoken language quite successfully. In order to extent this approach for classifying video content, a visual analog to words is constructed from signal-level visual features. A common bag-of-words approach is applied in order to represent video documents. Subsequently, support vector machines are trained to categorize the documents into known classes by using the proposed visual words. Experimental results indicating the classification performance are given and discussed.