Title :
A real time 1080P 30FPS Gaussian Mixture Modeling design for background subtraction and object extraction
Author :
Shuo-Wen Hsu ; Tian Sheuan Chang
Author_Institution :
Dept. of Electron. Eng. & Inst. of Electron., Nat. Chiao-Tung Univ., Hsinchu, Taiwan
Abstract :
The Gaussian Mixture Modeling (GMM) algorithm with connected component labeling as object extraction provide robust background subtraction but suffer from complexity, and large buffer or high bandwidth due to the frame level operations. For real time application needs, this paper proposed a block based GMM design for background subtraction with message passing between blocks to avoid performance drop. The corresponding parallel hardware design can reach real time 1080P@30fps and cost 164.82K gate-counts at 125MHz with 90nm process.
Keywords :
Gaussian processes; feature extraction; message passing; mixture models; object detection; GMM algorithm; Gaussian mixture modeling algorithm; background subtraction; block based GMM design; connected component labeling; frame level operations; frequency 125 MHz; message passing; object extraction; parallel hardware design; size 90 nm; Algorithm design and analysis; Bandwidth; Classification algorithms; Computer architecture; Hardware; Real-time systems; Robustness; Architecture; Background Subtraction; Gaussian Mixture Modeling;
Conference_Titel :
Consumer Electronics - Taiwan (ICCE-TW), 2014 IEEE International Conference on
Conference_Location :
Taipei
DOI :
10.1109/ICCE-TW.2014.6904050