Hierarchical video indexing and retrieval for subband-coded video

In this paper, we present a multiresolution approach for c-ideo indexing an d feature matching of subband-coded video databases. Four different scene-c hange detectors were tested; scene-change detection is applied only on the lowest subband for computational efficiency. Two kinds of scene changes, ab rupt and smoothly accumulated, mark the beginning of new scene segments. Th e index for each scene segment is the pair of the histograms of two represe ntative frames, the first and the last frame of the scene. Using the approa ch of query bg example, the index-matching algorithm takes a multiresolutio n approach by hierarchically comparing histograms at different resolutions. The search algorithm for the match between example query and its target sc ene segment starts from the coarsest resolution and moves to the next finer resolution until the finest resolution is reached. Experimental results ar e presented, and the proposed indexing technique appears to be promising fo r its speed and its inherent hierarchical search procedure.