| 摘要 |
PROBLEM TO BE SOLVED: To provide indexed timelines for use in video reproduction in video image processing. SOLUTION: Faces of speakers in a meeting or conference are automatically detected and facial images corresponding to each speaker are stored in a faces database. A timeline is created to graphically identify when each speaker is speaking during playback of a recording of the meeting. Instead of generically identifying each speaker in the timeline, a facial image is shown to identify each speaker associated with the timeline. COPYRIGHT: (C)2006,JPO&NCIPI
|
