Automatic Shot Boundary Detection and Classification of Indoor and Outdoor Scenes
Andrea Miene, Thorsten Hermes, George T. Ioannidis, Rahim Fathi, Otthein Herzog · 2002
Abstract. This paper describes our contribution to the TREC 2002 video analysis track. We participated in the shot detection task and in the feature extraction task (features indoors and outdoors). The shot detection approach is based on histogram differences and uses adaptive thresholds. Multiple detected shot boundaries that follow each other within a short temporal interval are grouped together and classified as a gradual change beginning with the first and ending with the last shot boundary in the interval. For the feature extraction task we examined whether it is possible to classify indoor and outdoor shots by their color distribution. In order to analyze the color distribution we use first order statistical features. The shots are classified into indoor and outdoor shots using a neural net.