Video Object Tracking Using Particle Filtering and Mean Shift

• Main Authors: Yu-cheng Huang, 黃昱郕
• Other Authors: Leu-Shing Lan
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/79466597224860146560

碩士 === 雲林科技大學 === 電子與資訊工程研究所 === 96 === Visual tracking of moving objects is an active research area in computer vision. Among the various schemes proposed, particle filtering and mean shift have both effectively demonstrated their success. Based on the usefulness of these methods, we are motivated to propose a new scheme that combines merits of both. In specific, we concentrate on the dynamic adjustment of kernel size and a modified measurement model. We report encouraging results under wide ranges of test conditions.