Compressive Acquisition of Dynamic Scenes
Compressive sensing (CS) is a new approach for the acquisition and recovery of sparse signals and images that enables sampling rates significantly below the classical Nyquist rate. Despite significant progress in the theory and methods of CS, little headway has been made in compressive video acquisition and recovery. Video CS is complicated by the ephemeral nature of dynamic events, which makes direct extensions of standard CS imaging architectures and signal models infeasible. In this paper, we develop a new framework for video CS for dynamic textured scenes that models the evolution of the scene as a linear dynamical system (LDS). This reduces the video recovery problem to first estimating the model parameters of the LDS from compressive measurements, from which the image frames are then reconstructed. We exploit the low-dimensional dynamic parameters (the state sequence) and high-dimensional static parameters (the observation matrix) of the LDS to devise a novel compressive measurement strategy that measures only the dynamic part of the scene at each instant and accumulates measurements over time to estimate the static parameters. This enables us to considerably lower the compressive measurement rate considerably. We validate our approach with a range of experiments including classification experiments that highlight the effectiveness of the proposed approach.
KeywordsCompressive Sense State Sequence Linear Dynamical System Video Model Dynamic Scene
Unable to display preview. Download preview PDF.
- 5.Chan, A.B., Vasconcelos, N.: Probabilistic kernels for the classification of auto-regressive visual processes. In: IEEE Conf. on Computer Vision and Pattern Recognition. pp. 846–851 (2005)Google Scholar
- 9.Fowler, J.: Compressive-projection principal component analysis. IEEE Transactions on Image Processing 18(10) (October 2009)Google Scholar
- 12.Park, J., Wakin, M.: A multiscale framework for compressive sensing of video. In: Picture Coding Symposium, pp. 197–200 (May 2009)Google Scholar
- 13.Péteri, R., Fazekas, S., Huiskes, M.: DynTex: A Comprehensive Database of Dynamic Textures (to appear, 2010), http://projects.cwi.nl/dyntex/
- 14.Saisan, P., Doretto, G., Wu, Y., Soatto, S.: Dynamic texture recognition. In: CVPR. vol. 2, pp. 58–63 (December 2001)Google Scholar
- 15.Turaga, P., Veeraraghavan, A., Chellappa, R.: Unsupervised view and rate invariant clustering of video sequences. CVIU 113(3), 353–371 (2009)Google Scholar
- 16.Vaswani, N.: Kalman filtered compressed sensing. In: ICIP (2008)Google Scholar
- 17.Vaswani, N., Lu, W.: Modified-CS: Modifying compressive sensing for problems with partially known support. In: Intl. Symposium on Information Theory (2009)Google Scholar
- 18.Veeraraghavan, A., Reddy, D., Raskar, R.: Coded strobing photography: Compressive sensing of high-speed periodic events. TPAMI (to appear)Google Scholar
- 19.Veeraraghavan, A., Roy-Chowdhury, A.K., Chellappa, R.: Matching shape sequences in video with applications in human movement analysis. TPAMI 27, 1896–1909 (2005)Google Scholar
- 21.Wakin, M., Laska, J., Duarte, M., Baron, D., Sarvotham, S., Takhar, D., Kelly, K., Baraniuk, R.: Compressive imaging for video representation and coding. In: Picture Coding Symposium (April 2006)Google Scholar