基于视差场校正和区域分割的中间视图像生成与内插方法  

An algorithm for intermediate view image creation and interpolation based on disparity field calibration and region segmentation

作　　者：骆艳[1] 安平[1] 张兆扬[1] 

机构地区：[1]上海大学通信与信息工程学院,上海200072

出　　处：《通信学报》2004年第10期127-133,共7页Journal on Communications

基　　金：国家自然科学基金资助项目(69972027;60202015);上海市重点学科建设基金资助项目(2001-44)

摘　　要：针对中间视图像生成依赖于精确稠密视差场估计以及遮挡区域检测的问题,提出了利用被编码传输的视差场进行中间视生成与内插算法,避免了复杂的视差场估计。本文算法将区域分割与视差场校正相结合,可对原有的误匹配视差进行校正。在中间视内插中,使用区域分割技术来对具有遮挡和模糊区域的内容进行判断,最后根据视差场的特性进行填充。实验结果表明生成的中间视图像具有良好的视觉效果。Since the intermediate view image creation have to depend on a perceive and dense disparity estimation and on the detection of occlusion, An intermediate view creation and interpolation algorithm were proposed by taking advantages of coded and transmitted disparity fields to avoid complex disparity estimation. This algorithm also combined the region segmentation and disparity calibration techniques together to calibrate those false matched disparity. What’s more, in the intermediate view creation, region segmentation technique was used to judge the contents of areas with occlusion and blur, and filled them according to the disparity characteristics. Experiment results show that constructed intermediate view images have favorite view effect.

关 键 词：图像处理 中间视生成 三维视频会议系统 立体视 区域分割 视差场校正 

分 类 号：TN91[电子电信—通信与信息系统]