A model change detection approach to dynamic scene modeling
AVSS '09 Proceedings of the 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance(2009)
摘要
In this work we propose a dynamic scene model to provide information about the presence of salient motion in the scene, and that could be used for focusing the attention of a pan/tilt/zoom camera, or for background modeling purposes. Rather than proposing a set of saliency detectors, we define what we mean by salient motion, and propose a precise model for it. Detecting salient motion becomes equivalent to detecting a model change. We derive optimal online procedures to solve this problem, which enable a very fast implementation. Promising results show that our model can effectively detect salient motion even in severely cluttered scenes, and while a camera is panning and tilting.
更多查看译文
关键词
salient motion detection,model change detection approach,model change detection,linear dynamical systems,ptz camera,traffic data extraction,traffic lights control,dynamic textures,pan-tilt-zoom camera,salient scene motion,object detection,increasing demand,focus-of-attention,dynamic scene modeling,background modeling,new framework,population evacuation,traffic issue,sequential generalized likelihood ratio,image motion analysis,mathematical model,noise measurement,change detection,noise,computational modeling,dynamics,linear dynamical system
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要