%0 Conference Paper %B Proceedings in the ICASSP'04 IEEE International Conference on Computer Vision %D 2005 %T Model of Object-Based Coding for Surveillance Video %A Yu,Yang %A David Doermann %X In this paper, we explore the model of potential savingsof object-based coding for surveillance video. Moving foreground objects in stationary camera surveillance video are detected by a background subtraction technique [3] and encoded with MPEG-4 object-based coding. Experiment results show that compared with frame-based coding, object-based coding can achieve significant savings which are dependent on the video content. We further model the relationship of compression efficiency and the number and size of video objects using statistical learning method. Simulations show that the model is representative. The model can be used to predict the savings of object-based coding and select the coding methods for surveillance video. %B Proceedings in the ICASSP'04 IEEE International Conference on Computer Vision %P 693 - 696 %8 2005/03// %G eng