中文摘要：

针对目前移动增强现实应用中仅能对少量目标、而不能满足移动用户对大量目标识别和跟踪要求的问题，提出一种基于二维码的新型移动增强现实系统．利用二维码海量内容编码技术和快速定位技术实现了对大量目标的识别，采用关键帧和预估第4对匹配点的方法实现对场景的快速跟踪，再通过手机端的SQLite数据容器存储3D模型实现了本地渲染，减少了从服务器下载3D模型的时延和对无线网络带宽的占用．实验结果表明，文中系统能对大量目标进行有效地识别和快速跟踪，且不依赖服务器实现虚实场景的融合显示．

英文摘要：

In this paper, a new mobile augmented reality system based on two-dimensional code is proposed. This system is designed to compensate the shortcoming existing in current mobile augmented reality applications that only a small number of objects can be identified and tracked in real- time, hence they can not be used in large-scale scenes for mobile users. Utilizing functions of mass content coding and rapid positioning of the two-dimensional code, the identification and tracking of objects in large-scale scenes are achieved. Additionally, to reduce the delay of downloading 3D models from the server, 3D models are stored in a local SQLite data container on mobile device and used for rendering locally. Experimental results show that the proposed method effectively identified a large number of objects and achieved fast tracking. Furthermore, the proposed method does not require server to participate in the fusion of virtual scenes and real scenes.