Semantic description of a video using representative frames | IEEE Conference Publication | IEEE Xplore