Here is the MPEG 7 Template provided originally by John R. Smith that will be used to return feature extraction results for evaluation and exchange. The confidence with which a feature has been detected should be entered in the "confidence=" field in the MPEG-7 descriptor. It is assumed that this number will take values in the range of 0.0 to 1.0. The use of "relevance =" field is left up to participating groups to experiment with. It will be ignored as far as the TREC 2002 Video Track evaluations are concerned. The MPEG-7 template assumes the common shot reference with shot IDs for the video track (i.e., shotxxxxxx, shotyyyyyy, shotzzzzzz)