An Efficient Neural Network Based Background Subtraction Method
The paper presents a neural network based segmentation method which can extract moving objects in video. This proposed neural network architecture is multilayer so as to match the complexity of the frames in a video stream and deal with the problems of segmentation. The neural network combines inputs that exploit spatio-temporal correlation among pixels. Each of these unit themselves produce imperfect results, but the neural network learns to combine their results for better overall segmentation, even though it is trained with noisy results from a simpler method. The proposed algorithm converges from an initial stage where all the pixels are considered to be part of the background to a stage where only the appropriate pixels are classified as background. Results are shown to demonstrate the efficacy of the method compared to a more memory intensive MoG method.
KeywordsBackground subtraction Neural networks Spatial–temporal correlation Backpropagation
Unable to display preview. Download preview PDF.
- Luque R.M., Domninguez E., Palomo E.J., Munoz J., “A Neural Network Approach for Video Object Segmentation in Traffic Surveillance”, ICIAR 2008, LNCS 5112, 151-158(2008).Google Scholar
- Luque R.M., Domninguez E., Palomo E.J., “A Dipolar Competitive Neural Network Video Object Segmentation”, IBERAMIA 2008, LNAI 5290, 103-112(2008).Google Scholar
- Culbrick Dubravko, Marques Oges,“Neural Network Approach to Background Modeling for Video Object Segmentation”, IEEE Transactions on Neural Networks, Vol. 18, No. 6, November 2007.Google Scholar
- Humpherys James, Hunter Andrew, “Multiple object tracking using a neural cost function”, Image Vision Computing, 417-424, 27(2009).Google Scholar
- Owens Jonathan, Hunter Andrew, Fletcher Eric, “A Fast Model Free Morphology Based Object Tracking Algorithm”, BMVC 2002.Google Scholar
- VISOR, http://www.openvisor.org
- Fabian Wauthier, http://www.cs.berkeley.edu/~flw/tracker