Abstract
This work is focused on the assessment of the use of GPU computation in dynamic texture segmentation under the mixture of dynamic textures (MDT) model. In this generative video model, the observed texture is a time-varying process commanded by a hidden state process. The use of mixtures in this model allows simultaneously handling of different visual processes. Nowadays, the use of GPU computing is growing in high-performance applications, but the adaptation of existing algorithms in such a way as to obtain a benefit from its use is not an easy task. In this paper, we made two implementations, one in CPU and the other in GPU, of a known segmentation algorithm based on MDT. In the MDT algorithm, there is a matrix inversion process that is highly demanding in terms of computing power. We make a comparison between the gain in performance obtained by porting to GPU this matrix inversion process and the gain obtained by porting to GPU the whole MDT segmentation process. We also study real-time motion segmentation performance by separating the learning part of the algorithm from the segmentation part, leaving the learning stage as an off-line process and keeping the segmentation as an online process. The results of performance analyses allow us to decide the cases in which the full GPU implementation of the motion segmentation process is worthwhile.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Berthold, K.P.H., Schunck, B.G.: Determining optical flow. In: Artificial Intelligence, vol. 17, no. 1–3, pp. 185–203 (1981)
Lucas, B.D., Kanade, T.: An iterative image registration technique with an application to stereo vision, in IJCAI81,1981, pp. 674–679 (1981)
Doretto, G.: Dynamic textures: Modeling, learning, synthesis, animation, segmentation, and recognition, Thesis (Ph.D.), University of California, Los Angeles (2013)
Chan, A.B., Vasconcelos, N.: Modeling, clustering, and segmenting video with mixtures of dynamic textures, PAMI, vol. 30 (2008)
Chan, A.B., Vasconcelos, N.: Variational layered dynamic textures, in CVPR, pp. 1062–1069 (2009)
Roweis, S., Ghahramani, Z: A unifying review of linear Gaussian models. Neural. Comput. 11(2):305–345 (1999)
Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. Roy. Statist. Soc. Ser. B. Math. 39(1), 1–38 (1977)
Chan, A.: Synthetic dynamic texture segmentation database. http://www.svcl.ucsd.edu/projects/motiondytex/db/dytex_synthdb.zip (2009)
Hubert, L., Arabie, P.: Comparing partitions (1985)
Huang, M.-Y., Wei, S.-C., Huang, B., Chang, Y.-L.: Accelerating the Kalman Filter on a GPU, ICPADS (2011)
Bouckaert, R.: Matrix inverse with Cuda and CUBLAS. http://www.cs.waikato.ac.nz/remco/
Ltaief, H., Tomov, S., Nath, R., Du, P., Dongarra, J.: A scalable high performant cholesky factorization for multicore with GPU accelerators,” in VECPAR (2010)
NVIDIA, Cuda cublas library, Jan 2012, Version 4.1
Chan, A.B., Vasconcelos, N.: Layered dynamic textures. IEEE Transactions on Pattern Analysis and Machine Intelligence 31, 1862–1879 (2009)
Chan, A.B., Vasconcelos, N.: Probabilistic kernels for the classification of auto-regressive visual processes. IEEE in Computer Vision and Pattern Recognition, 2005. IEEE Computer Society Conference on CVPR 2005. 1, 846–851 (2005)
Chan, A.B., Vasconcelos, N.: Classification and retrieval of traffic video using auto-regressive stochastic processes. Proceedings of 2005 IEEE Intelligent Vehicles Symposium, Las Vegas (2005)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Gómez Fernández, F., Buemi, M.E., Rodríguez, J.M. et al. Performance of dynamic texture segmentation using GPU. J Real-Time Image Proc 11, 375–383 (2016). https://doi.org/10.1007/s11554-013-0368-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11554-013-0368-8