JACIII Vol.18 No.4 pp. 632-647
doi: 10.20965/jaciii.2014.p0632


Estimation of Locations of Densely Distributed Subjects Using NMF with Nonpixel Information

Yusuke Kubo, Masao Kubo, Hiroshi Sato,
and Akira Namatame

National Defense Academy of Japan, 1-10-20 Hashirimizu, Yokosuka, Kanagawa 239-8686, Japan

July 21, 2013
March 12, 2014
July 20, 2014
nonnegative matrix factorization (NMF), collective intelligence, social sensing, social mining, GIS
We propose a method that uses a large number of digital photographs to produce highly accurate estimates of the locations of subjects that have attracted a crowd’s attention. Recently, a very active area of research has been to use humans as sensors in realworld observations that require a large amount of data. Some of these studies have attempted to produce real-time estimates of the subjects that are attracting a crowd’s attention by quickly collecting a large number of photographs. These studies are based on the assumption that, when photographers encounter interesting events, they take pictures. Some of the proposed methods realize high availability by using only photographing information, which includes information about location and azimuth of the camera and it is automatically embedded into photograph. Since this data is very small compared to that of the pixel information, the load on the communication infrastructure is reduced. However, there are problems with the accuracy when there are many attractive subjects in a small region, and they cannot be found with traditional methods that use a sequential search strategy. The proposed method overcomes this problem by applying nonnegative matrix factorization (NMF) to the estimated location of each subject. We verified the effectiveness of this by computational experiments and an experiment under a realistic environment.
Cite this article as:
Y. Kubo, M. Kubo, H. Sato, and A. Namatame, “Estimation of Locations of Densely Distributed Subjects Using NMF with Nonpixel Information,” J. Adv. Comput. Intell. Intell. Inform., Vol.18 No.4, pp. 632-647, 2014.
Data files:
  1. [1] T. Sakaki, M. Ozaki, and Y. Matsuo, “Earthquake shakes Twitter users: real-time event detection by social sensors,” Proc. of the 19th Int. Conf. on World Wide Web, pp. 851-860, ACM, 2010.
  2. [2] E. Aramaki, S.Maskawa, and M.Morita, “Twitter Catches The Flu: Detecting Influenza Epidemics using Twitter,” Proc. EMNLP 2011, pp. 1568-1576, 2012.
  3. [3] M. Paul and M. Dredze, “You Are What You Tweet: Analyzing Twitter for Public Health,” Proc. ICWSM 2011, 2011.
  4. [4] S. Kurihara, Y. Okada, and M. Numao, “SIR-Extended Information Diffusion Model of False Rumor and its Prevention Strategy for Twitter,” Proc. WEIN 2013, pp. 114-128, 2013.
  5. [5] D. Crandall, L. Backstrom, D. Huttenlocher, and J. Kleinberg, “Mapping the world’s photos,” Proc. WWW 2009, pp. 761-770, 2009.
  6. [6] Z. Yin, L. Cao, J. Han, C. Zhai, and T. Huang, “Geographical topic discovery and comparison,” Proc. WWW 2011, pp. 247-256, 2011.
  7. [7] M. Kumano, M. Koseki, K. Ono, and M. Kimura, “Extracing Hot Photo-spots from Geotagged Photographs with Timestamps,” IPSJ J., Vol.5, No.3, pp. 41-53, 2012.
  8. [8] M. Kubo, S. Nakayama, and H. Sato, “Collective Discovery of Geographic Locations of Frequently Photographed Objects Only using the Metadata of Digital Photographs,” Procedia Computer Science, Vol.10, pp. 625-633, 2012.
  9. [9] Y. Kubo, M. Kubo, H. Sato, M. Hirano, and A. Namatame, “Understanding Geographic Attentions of Crowd from Photographing Information,” J of Advanced Computational Intelligence and Intelligent Informatics (JACIII), Vol.17, No.6, pp. 890-903, 2013.
  10. [10] A. Abe, T. Sasaki, and N. Odashima, “Development and Operational Evaluation of the Groupware Based on Geographical Location Information for Local Community Activities,” IPSJ J., Vol.45, No.1, pp. 155-163, 2004.
  11. [11] D. Lee and H. Seug, “Learning the parts of objects by non-negative matrix factorization,” Nature, Vol.401, pp. 788-791, 1999.
  12. [12] P. Smaragdis and J. Brown, “Non-negative matrix factorization for polyphonic music transcription,” Proc. WASPAA 2003, pp. 177-180, 2003.
  13. [13] X. Wei, L. Xin, and G. Yihong, “Document clustering based on nonnegative matrix factorization,” Proc. ACM SIGIR 2003, pp. 267-273, 2003.
  14. [14] H. Fujita, M. Arikawa, and K. Okamura, “Three-Dimensional Real Space Mapping of Photographs with Precise Spatial Metadata,” IEICE Trans. A, Vol.J87-A, No.1, pp. 120-131, 2004.
  15. [15] M. Shirai, M. Hirota, S. Yokoyama, N. Fukuta, and H. Ishikawa, “A System for Reproducing Multi-viewpoint Landmarks using Geotagged Photos,” Proc. DEIM Forum 2012, 2012.
  16. [16] D. Lee and H. Seug, “Algorithms for nonnegative matrix factorization,” Proc. NIPS 2000, pp. 556-562, 2000.
  17. [17] C. Fevotte, N. Bertin, and J.-L. Durrieu, “Nonnegative matrix factorization with the Itakura-Saito divergence: With application application to music analysis,” NECO, Vol.21, No.3, pp. 793-830, 2009.
  18. [18] H. Sawada, “Nonnegative Matrix Factorization and Its Applications to Data/Signal Analysis,” IEICE J., Vol.95, No.9, pp. 829-833, 2012.
  19. [19] Standard of the Camera & Imaging Products Association, “Exchangeable image file format for digital still camera: Exif Version 2.3,” 2010.
  20. [20] C. M. Bishop et al., “Pattern Recognition and Machine Learning,” pp. 541-542, Springer, 2006.

*This site is desgined based on HTML5 and CSS3 for modern browsers, e.g. Chrome, Firefox, Safari, Edge, Opera.

Last updated on Jul. 23, 2024