摘要
基于单幅图像的三维重建,需要的信息少,避免了基于多幅图像三维重建时图像匹配的难点,成为基于图像三维重建研究的一个热点。物体的深度值的计算是单幅图像三维重建的研究难点,针对此难点,利用规则物体的几何性质,对单幅未标定图像进行三维重建。首先利用消失点进行摄像机标定,然后手动获得需要的最少二维点,通过标定矩阵和物体几何性质计算出这些二维点的三维坐标,进而得到物体精确重建需要的所有信息,最后利用OpenGL实现三维重建和纹理映射。实验表明,此方法可恢复出目标物体的精确三维结构,能够满足一般虚拟现实的需要。
3D reconstruction from a single image has been an active research topic for many years, which needs less information and avoids the matching difficulty of the 3D reconstruction of multiple images. The calculation of objects＂ depth value is a hard work for reconstruction from a single image. In order to figure out this problem, some objects are rebuilt by using the geometric properties of rule geometry from a single image. Firstly, the camera is calibrated by using the vanishing points. Secondly, the least two-dimensional points needed are captured manually, and the geometric properties of the objects are calculated by using calibration matrix and geometric constraints. Finally, the 3D reconstruction and texture mapping is imple- mented by OpenGL. The experiments demonstrate that this approach can recover the entire three-dimen- sional structure of the object accurately and meet the needs of the general virtual reality.
出处
《青岛大学学报（自然科学版）》
CAS
2013年第1期55-59,65,共6页
Journal of Qingdao University(Natural Science Edition)
基金
国家自然科学基金项目(61170106)
青岛市科技计划项目基金(06-2-2-2-JC)
关键词
单幅未标定图像
消失点
相机标定
物体几何性质
三维重建
single uncalibrated image vanishing points camera calibration geometric properties of ob-jects 3D reconstruction
