A New Object-Based Video Inpainting Method

• Main Authors: Chien-Chih Lin, 林建志
• Other Authors: Jin-Jang Leou
• Format: Others
• Language: en_US
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/11764600536040784999

碩士 === 國立中正大學 === 資訊工程所 === 98 === Vidoe inpainting is to remove unwants objects or reconstruct damaged parts of an image automatically, which is an ill-posed problem in video processing. In this study, a video inpainting approach using edge-guided belief propagation (BP) and graph cuts on spatiotemporal domain is proposed. The proposed video inpainting approach contains five stages, i.e., background/foreground segmentation, and foreground objects store in a database, dissimilarity measure determines the most match foreground object, video inpainting using priority-BP optimization, and video reconstruction. Based on experimental results obtained in this study, as compared with two comparison methods, the proposed approach provides the better video inpainting results