增强现实中的虚实遮挡处理综述  

Survey on Occlusion Handling in Augmented Reality

作　　者：徐维鹏[1] 王涌天[1] 刘越[1] 翁冬冬[1] 

机构地区：[1]北京理工大学光电学院,北京100081

出　　处：《计算机辅助设计与图形学学报》2013年第11期1635-1642,共8页Journal of Computer-Aided Design & Computer Graphics

基　　金：国家自然科学基金(61072096);国家科技重大专项(2012ZX03002004);国家科技支撑计划(2012BAH64F04)

摘　　要：虚实遮挡处理问题的研究对于增强现实系统的完善和实用化起到至关重要的作用,是增强现实应用中迫切需要解决的关键问题.由于受到图像前景分割、三维深度提取等相关关键技术的制约,目前国内外尚未出现完整的普适性解决方案.文中对增强现实中虚实遮挡处理问题的关键理论和技术进行了回顾和总结.首先对虚实遮挡处理问题及处理流程进行了描述;之后分析了国内外研究现状,对现有的基于建模、基于深度以及基于图像分析等3类处理方法进行了讨论,给出了每类技术的优缺点及适用范围,并总结了2种典型的改进方法;最后探讨了该领域的发展方向.Occlusion handling plays a very important role in Augmented Reality system, which is a key issue to be addressed. Although this problem has been extensively studied, there is still no complete and general solution due to the limitations on related key technologies such as foreground segmentation and depth extraction. This paper presents a technical review on recent research progresses regarding occlusion handling in Augmented Reality. The paper contains four parts. Firstly, the processing scenario of mutual occlusion in Augmented Reality is described. Secondly, three types of the state-of- the-art techniques for occlusion handling are discussed, including model based approaches, depth based approaches and image based approaches. In the third part, performance and application scope of each technique are discussed, and two typical improved approaches are concluded. Finally, the research trends are discussed.

关 键 词：增强现实 虚实遮挡 三维重建 深度提取 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]