Method and system for producing relevance sorted video summary are provided herein. The method may include: obtaining a source video containing a plurality of source objects; receiving features descriptive of at least some of the source objects; clustering the source objects into clusters, each cluster including source objects that are similar in respect to one of the features or a combination of the features; obtaining relevance level of the clustered source objects, respectively; generating synopsis objects by sampling respective clustered source objects; and generating a synopsis video having an overall play time shorter than the overall play time of the source video by determining a play time for each of the synopsis objects based at least partially on the respective relevance level.