{"references": ["A. G. Money, H. Agius, \"Video summarization: a conceptual framework\nand survey of the state of the art,\" Journal of Visual Communication and\nImage Representation, Vol. 19, No. 2, pp. 121-143, 2008.", "Ajmal, Muhammad and Ashraf, Muhammad Husnain and Shakir,\nMuhammad and Abbas, Yasir and Shah, Faiz Ali, \"Video Summarization:\nTechniques and Classification,\" Proceedings of the 2012 International\nConference on Computer Vision and Graphics, pp. 1\u201313, 2012.", "B. T. Troung, S. Venkatesh,\"Video abstraction: a systematic review\nand classification,\" ACM Transactions Multimedia Computing,\nCommunications and Applications, Vol. 3, No. 1, 2007.", "M. Furini, F. Geraci, and M. Montangero, \"VISTO: Visual STOryboard\nfor web video browsing,\" CIVR, pp.635-641, 2007.", "Z. Li, G. M. Schuster, and A. K. Katsaggelos, \"MINMAX optimal video\nsummarization,\" IEEE Trans Circuits Syst. Video Technol., vol.15, no.10,\npp.1245-1256, 2005.", "C. Panagiotakis, A. Doulamis, and G. Tziritas, \"Equivalent key frames\nselection based on iso-content principles,\" IEEE Trans. Circuits Syst.\nVideo Technol., vol.19, no.3, pp.447-451, 2009.", "G. Guan, Z. Wang, S. Lu, J. D. Deng, and D. D. Feng, \"Keypoints-based\nkeyframe selection,\" IEEE Trans. Circuits Syst. Video Technol., vol.23,\nno.4, 2013.", "S. E. D. Avila, A. B. P, Lopes, L. J. Antonio, and A. d. A. Araujo,\n\"VSUMM: a mechanism designed to produce static video summaries\nand novel evaluation method,\" Pattern Recognition Letter, vol.32 (1),\npp.56-68, 2011.", "N. Ejaz, T. B. Tariq, and S. W. Balik, \"Adaptive key frame extraction for\nvideo summarization using an aggregating mechanism,\" Journal of Visual\nCommunication and Image Representation, vol.23, pp.1031-1040, 2012.\n[10] N. D. Doulamis, A. D. Doulamis, Y. Avrithis, and S. D. Kollias, \"A\nstochastic framework for optimal frame extraction from MPEG video\ndatabases,\" Comput. Visi. Image Understand., vol.75, no.1-2, pp.3-24,\n1999.\n[11] A. Nagasaka and Y. Tanaka, \"Automatic video indexing and full-video\nsearch for object appearances,\" in Visual Database Systems II, 1992.\n[12] Y. Zhuang, Y. Rui, T. Huang, and S. Mehrotra, \"Adaptive key frame\nextraction using unsupervised clustering,\" in Proc. IEEE Int. Image\nProcess., pp.866-870, 1998.\n[13] P. Mundur, Y. Rao, and Y. Yesha, \"Keyframe-based video summarization\nusing Delaunay clustering,\" International Journal on Digital Libraries\n(IJDL) 6(2), pp.219-232, 2006.\n[14] M. Furini, F. Geraci, M. Montangero, and M. Pellegrini, \"STIMO: STIll\nand MOving video storyboard for the web scenario,\" Multimedia Tools\nand Applications, vol.46, no.1, pp.47-69, 2010.\n[15] J. Almeida, N. J. Leite, and Ricardo da S. Torres, \"VISON: VIdeo\nSummarization for ONline applications,\" Pattern Recogn. Lett. 33, 4\n(March 2012), pp. 397-409, 2012.\n[16] Z. Zhao and A. Elgammal, \"Information theoretic key frame selection\nfor action recognition,\" in Proc. Of British machine vision, pp.1-10, 2008.\n[17] T. Liu, H. J. Zhang, and F. Qi, \"A novel video key-frame-extraction\nalgorithm based on perceived motion energy model,\" IEEE Trans. Circuits\nSyst. Video Technol., vol.13, no.10, pp.1006-1013, 2013.\n[18] B. Fauvet, P. Bouthemy, P. Gros, and F. Spindler, \"A geometrical\nkey-frame selection method exploiting dominant motion estimation in\nvideo,\" in Proc. CIVR, 2004.\n[19] W. Wolf, \"Key frame selection by motion analysis,\" in Proc. IEEE Int.\nConf. Acoust., Speech, and Signal Proc., 1996.\n[20] W. Barhoumi and E. Zagrouba, \"On-the-fly extraction of key frames for\nefficient video summarization,\" AASRI Conference on Intelligent Systems\nand Control, vol.4, pp.78-84, 2013.\n[21] N. Ejaz, I. Mehmood, and S. W. Baik, \"Efficient visual attention based\nframework for extracting key frames from videos,\" Signal Processing:\nImage Communication, vol.28(1), pp.34-44, 2013.\n[22] H. Zhang, J.Wu, D. Zhong, and S. W. Smoliar, \"An integrated system for\ncontent-based video retrieval and browsing,\" Pattern Recognition, vol.30,\nno.4, pp.643-658, 1997.\n[23] K. Torkkoa, \"Feature extraction by non-parametric mutual information,\"\nJ. Machine Learning Research, vol.3, pp.1415-1438, 2003.\n[24] J. Sainui and M. Sugiyama, \"Direct approximation of quadratic mutual\ninformation and its application to dependence-maximization clustering,\"\nIEICE Trans. Inf. & Syst., vol.E96-D, no.10, pp.2282-2285, 2013.\n[25] J. Sainui and M. Sugiyama, \"Minimum dependency key frames selection\nvia quadratic mutual information \", The 10th International Conference on\nDigital Information Management (ICDIM2015), pp. 148-153, 2015.\n[26] M.J. Swain and D.H. Ballard, \"Color indexing\", International Journal\nof Computer Vision, 7 (11), pp. 11-32, 1991.\n[27] The Open Video Project (Online). Available: http://www.open-video.org\n(Accessed on 27/10/2015).\n[28] Video SUMMarization (Online). Available:\nhttps://sites.google.com/site/vsummsite/download (Accessed on\n27/10/2015).\n[29] (Online). Available: http://www.liv.ic.unicamp.br/ jurandy/vison/VISON\nSummary.zip (Accessed on 16/05/2016).\n[30] (Online). Available: http://www.liv.ic.unicamp.br/\u223cjurandy/summaries\n(Accessed on 16/05/2016)."]}

As a rapid growth of digital videos and data communications, video summarization that provides a shorter version of the video for fast video browsing and retrieval is necessary. Key frame extraction is one of the mechanisms to generate video summary. In general, the extracted key frames should both represent the entire video content and contain minimum redundancy. However, most of the existing approaches heuristically select key frames; hence, the selected key frames may not be the most different frames and/or not cover the entire content of a video. In this paper, we propose a method of video summarization which provides the reasonable objective functions for selecting key frames. In particular, we apply a statistical dependency measure called quadratic mutual informaion as our objective functions for maximizing the coverage of the entire video content as well as minimizing the redundancy among selected key frames. The proposed key frame extraction algorithm finds key frames as an optimization problem. Through experiments, we demonstrate the success of the proposed video summarization approach that produces video summary with better coverage of the entire video content while less redundancy among key frames comparing to the state-of-the-art approaches.