In this work, a user-centered approach has been the basis for generation of the personalized video summaries. Primarily, the video experts score and annotate the video frames during the enrichment phase. Afterwards, the frames scores for different video segments will be updated based on the captured end-users (different with video experts) priorit ies towards existing video scenes. Eventually, based on the predefined skimming t ime, the highest scored video frames will be extracted to be included into the personalized video summaries. In order to evaluate the effect iveness of our proposed model, we have compared the video summaries generated by our system against the results from 4 other summarization tools using different modalities.