Automated high-level movie segmentation for advanced video-retrieval systems

We present a newly developed strategy for automatically segmenting movies i nto logical story units. A logical story unit can be understood as an appro ximation of a mol ie episode, which is a high-level temporal movie segment, characterized either by a single event (dialog, action scene, etc.) or by several events taking place in parallel. Since ne consider a whole el ent a nd not a single shot to be the most natural retrieval unit for the movie ca tegory of video programs, the proposed segmentation is the crucial first st ep toward a concise and comprehensive content-based movie representation fo r browsing and retrieval purposes. The automation aspect is becoming increa singly important with the rising amount of information to be processed in v ideo archives of the future. The segmentation process is designed to work o n MPEG-DC sequences, where we have taken into account that at least a parti al decoding is required for performing content-based operations on MPEG com pressed video streams. The proposed technique allows for carrying out the s egmentation procedure in a single pass through a video sequence.