Title :
Gaussian mixture model for video compressive sensing
Author :
Jianbo Yang ; Xin Yuan ; Xuejun Liao ; Llull, Patrick ; Sapiro, Guillermo ; Brady, David J. ; Carin, Lawrence
Author_Institution :
Electr. & Comput. Eng. Dept., Duke Univ., Durham, NC, USA
Abstract :
A Gaussian Mixture Model (GMM)-based algorithm is proposed for video reconstruction from temporal compressed measurements. The GMM is used to model spatio-temporal video patches, and the reconstruction can be efficiently computed based on analytic expressions. The developed GMM reconstruction method benefits from online adaptive learning and parallel computation. We demonstrate the efficacy of the proposed GMM with videos reconstructed from simulated compressive video measurements and from a real compressive video camera.
Keywords :
Gaussian processes; compressed sensing; data compression; learning (artificial intelligence); mixture models; parallel processing; signal reconstruction; video cameras; video coding; GMM reconstruction method; GMM-based algorithm; Gaussian mixture model; analytic expressions; compressive video camera; compressive video measurements; online adaptive learning; parallel computation; spatiotemporal video patches; temporal compressed measurements; video compressive sensing; video reconstruction; Cameras; Compressed sensing; Image coding; Image reconstruction; Sensors; Training; Compressive sensing; Gaussian mixture model; coded aperture compressive temporal imaging;
Conference_Titel :
Image Processing (ICIP), 2013 20th IEEE International Conference on
Conference_Location :
Melbourne, VIC
DOI :
10.1109/ICIP.2013.6738005
