文献类型：会议论文

英文题名：Effective Approach to Camera Zoom Detection Based on Visual Attention

作者：Huang Jing-hua[1];Yang Yan-song[1]

第一作者：黄静华

通讯作者：Huang, JH[1]

机构：[1]Beijing Union Univ, Inst Informat Technol, Beijing 100101, Peoples R China

第一机构：北京联合大学智慧城市学院

通讯机构：[1]corresponding author), Beijing Union Univ, Inst Informat Technol, Beijing 100101, Peoples R China.|[1141734]北京联合大学智慧城市学院;[11417]北京联合大学;

会议论文集：9th International Conference on Signal Processing

会议日期：OCT 26-29, 2008

会议地点：Beijing, PEOPLES R CHINA

语种：英文

外文关键词：Behavioral research - Cameras - Vectors

摘要：An effective approach based on geometric constraint of the motion vector field is proposed to detect camera zoom operation. The approach exploits the valid motion vectors to estimate the center of focus (COF) for zoom, and then calculates the average distance from the COF to valid motion vectors to identify zoom operation. To reduce the influence of object motion, Visual attention mechanism is employed to select valid motion vectors. Experimental results show that the proposed approach achieves a good performance.