In this paper diversity visualization of video content posted on Social Media is efficiently performed, by proposing an unsupervised intelligent video analysis scheme. The proposed scheme assumes several different videos, posted by several different social media users. Its aim is to provide an overall compact view of the diverse video content people share. Similarly, it is like providing a summary of the total posted visual information (for a specific time instance or interval), so that users can take an idea of what is happening outside their micro-world. Towards this direction, each video is analyzed and key-frames are extracted based on a correlation measure and a social computing algorithm. The final summary is created by extracting the most uncorrelated frames among all key-frames, so that the diversity of the visualized content is kept. Experimental results are presented, to denote the full potential of the proposed scheme, its advantages as well as important issues for future work.