Multiresolution Techniques in Video Shot Boundary Detection. A step towards high level Video Segmentation

      Аннотация: Video abstraction, high level video segmentation, video indexing, video summarization and video retrieval needs a basic step of detection of shot boundaries. Therefore detection of shot boundaries is prerequisite for revealing content of higher level video structure. There are various applications where shot boundary detection automation can be used such as Geographical Information System (GIS), restoration of video, multimedia news, digital libraries, telelearning, interactive display. There are some issues that still need to be addressed by the researchers to resolve. The disturbance caused by the illumination change, detection of gradual and abrupt transitions are the major challenges in the shot boundary detection. It is but difficult to implement such algorithm which is insensitive to illumination change disturbance and can detect the shot boundaries of all types. In this book we focus on illumination removal and detection of abrupt transition.The results are evaluated using performance metrics Recall, Precision and F1 measure.

Ключевые слова: Radon Transform, recall, Shot Boundary Detection, Video, wavelet transform, abrupt transition, ridgelet transform, cosine transform, precision. shot, cut transition, video sementation