Abstract
Video classification and clustering are key techniques in multimedia applications such as video segmentation and recognition. This paper investigates the application of incremental manifold learning algorithms to directly learn nonlinear relationships among video frames. Video frame classification and clustering are performed to the projected data in an intrinsic latent space. This approach has avoided partitioning video frames into arbitrary groups. It works even when the input video frames are under-sampled or unevenly distributed. Experiments show that video classification and clustering give better results in the latent space than in the original high dimensional space.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
- 3.
Available at http://www.cs.toronto.edu/~roweis/data.html.
- 4.
Silhouette value is a measure of how similar a data point is to data points in its own cluster versus data points in other clusters. Let \(a_i\) denote the average distance of the i-th point to all other points in the same cluster. Let \(b_i\) denote the minimum of average distances of the i-th point to all points in other clusters, that is, the average of distances to all points in the next closest cluster. The i-th point’s silhouette value is defined as \(s_i = (b_i - a_i)/\max (a_i, b_i)\). Silhouette values range from \({-}1\) to +1.
References
Law, M.H.C., Jain, A.K.: Incremental nonlinear dimensionality reduction by manifold learning. IEEE Trans. Pattern Anal. Mach. Intell. 28(3), 377–391 (2006)
Lee, K.C., Kriegman, D.: Online learning of probabilistic appearance manifolds for video-based recognition and tracking. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Washington, DC, vol. 1, pp. 852–859 (2005)
Roach, M., Mason, J., Xu, L., Stentiford, F.: Recent trends in video analysis : a taxonomy of video classification problems. In: Proceedings of the International Conference on Internet and Multimedia Systems and Applications, Honolulu, HI, pp. 864–871, August 2003
Seung, H.S., Lee, D.D.: The manifold ways of perception. Science 290(5500), 2268–2269 (2000)
Tenenbaum, J.B., de Silva, V., Langford, J.C.: A global geometric framework for nonlinear dimensionality reduction. Science 290, 2319–2323 (2000)
Yang, L.: Building k-connected neighborhood graphs for isometric data embedding. IEEE Trans. Pattern Anal. Mach. Intell. 28(5), 827–831 (2006)
Yeo, B.L., Liu, B.: Rapid scene analysis on compressed videos. IEEE Trans. Circ. Syst. Video Technol. 5(6), 533–544 (1995)
Yeung, M., Yeo, B.L.: Time-constrained clustering for segmentation of video into story units. In: Proceedings of the International Conference on Pattern Recognition, Vienna, Austria, pp. 375–380, August 1996
Yeung, M., Yeo, B.L., Liu, B.: Extracting story units from long programs for video browsing and navigation. In: Proceedings of the International Conference on Multimedia Computing and Systems, Hiroshima, Japan, pp. 296–305, June 1996
Zhang, H.J., Low, C.Y., Smoliar, S.W.: Video parsing and browsing using compressed data. Multimedia Tools Appl. 1(1), 89–111 (1995)
Zhao, D., Yang, L.: Incremental isometric embedding of high dimensional data using connected neighborhood graphs. IEEE Trans. Pattern Anal. Mach. Intell. 31(1), 86–98 (2009)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Yang, L., Wang, X. (2016). Online Appearance Manifold Learning for Video Classification and Clustering. In: Gervasi, O., et al. Computational Science and Its Applications – ICCSA 2016. ICCSA 2016. Lecture Notes in Computer Science(), vol 9787. Springer, Cham. https://doi.org/10.1007/978-3-319-42108-7_43
Download citation
DOI: https://doi.org/10.1007/978-3-319-42108-7_43
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-42107-0
Online ISBN: 978-3-319-42108-7
eBook Packages: Computer ScienceComputer Science (R0)