摘要
自标定方法可以得到相机系统的内部参数和/或外部参数,而不对场景结构作假设或作很少假设。本文提出一种新的自标定方法,用于投影相机建模系统。一个关键点在于投影仪既可以作为主动光源使用,也可以当做一个虚拟相机使用。首先,使用结构光方法从不同角度投影并拍摄一组图像。然后,通过将空间约束关系转换为内参约束,完成对系统的自标定。最后,利用估计的参数进行3D建模。实验结果验证了本文方法的可行性。
With no or few assumptions about the struclure of scene, self-calibration methods try to find the intrinsic parameters and/or extrinsic parameters of the system. This paper proposes a new self-calibration method for a projector-camera based 3D modeling system. The key enabling observation is that a projector can be used as an active light source, and be conceptually regarded as a camera. Firtly, we captore a few image sequences under three different hght position using structured-light method. Taking a simple constraint into account, we can then achieve a self-calibration of the system. Finally, we use the estimated parameters ta build a 3D model. The experimental resuhs show the effectiveness of this approach.
出处
《科技通报》
北大核心
2012年第7期130-134,共5页
Bulletin of Science and Technology
基金
国家863课题研究项目(编号:2006AA01Z307)
关键词
投影相机系统
自标定
3D建模
本质矩阵
projector-camera system
self-calibration
3D modeling
essential matrix