Title: Global motion estimation based on the reliability analysis of motion vectors
Abstract: Global motion estimation of videos depends on the image motion vector field,but unreliable motion vectors affect the estimation results. A global motion estimation algorithm was developed based on the reliability analysis of motion vectors to improve the estimation accuracy. The reliability of the current frame's motion vector field is first analyzed,with each motion vector then classified using support vector machine method into reliable one and unreliable one according to the block's matching features. The reliable motion vectors are used to construct a linear parameter model with the parameters determined by iterative linear least-square estimation. Experimental results show that the algorithm effectively eliminates the effect of regions with unreliable motion vectors and improves the estimation result,even if the conventional method fails.
Publication Year: 2010
Publication Date: 2010-01-01
Language: en
Type: article
Access and Citation
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot