Feature Tracking for Cut Detection

Recently, investigation into shot boundary detection schemes has gathered much momentum [20-29]. Cut detection is seemingly easily solved by an elementary statistical examination of inter-frame characteristics; however a truly accurate and generalized cut detection algorithm still eludes researchers. Reliable shot boundary detection forms the cornerstone for video segmentation applications as shots are considered to be the elementary building blocks that form complete video sequences. Applications such as video abstraction, video retrieval and higher contextual segmentation all presuppose an accurate solution to the shot boundary detection problem [8, 17, 18, 19]. Automatic recovery of these shot boundaries is an imperative primary step, and accuracy is essential. Shot transitions can be classified into four classes based on the 2D image transformations applied during transition production [20]. • Identity Class: Neither of the two shots involved are modified, and no additional edit frames are added. Only hard cuts qualify for this class. • Spatial Class: Some spatial transformations are applied to the two shots involved. Examples are wipe, page turn, slide, and iris effects. • Chromatic Class: Some color space transformations are applied to the two shots involved. Examples are fades and dissolve effects. • Spatio-Chromatic Class: Some spatial as well as some color space transformations are applied to the two shots involved. All morphing effects fall into this category.