Three-dimensional cinematography with control object of unknown shape.
Three-dimensional cinematography with control object of unknown shape.
复制标题
具有未知形状的控制对象的三维电影摄影。
DOI：
--
复制
发表时间：
1981
影响因子：
2.4
通讯作者：
J. A. Miller
中科院分区：
文献类型：
--
作者：
J. Dapena;E. Harman;J. A. Miller
开发了一种重建三维(3D)运动的技术，该技术涉及简单的拍摄过程，但允许在大对象体积中推导坐标。内部相机参数是根据两个校准十字的胶片图像的测量来计算的，而外部相机参数是从形状未知但至少有一个已知长度的控制对象中的点的胶片图像来计算的。控制对象包括发生活动的体积，由放置在未知位置的一系列杆子组成，每个杆子携带两个目标。根据相机的内部和外部参数，以及从两个相机的胶片中点的图像的位置，可以计算点的3D坐标。大体积(5m×5m×1.5m)三点坐标的均方根误差分别为15 mm、13 mm、13 mm和6 mm，长度的相对误差分别为0.5%、0.7%和0.5%。
A technique for reconstruction of three-dimensional (3D) motion which involves a simple filming procedure but allows the deduction of coordinates in large object volumes was developed. Internal camera parameters are calculated from measurements of the film images of two calibrated crosses while external camera parameters are calculated from the film images of points in a control object of unknown shape but at least one known length. The control object, which includes the volume in which the activity is to take place, is formed by a series of poles placed at unknown locations, each carrying two targets. From the internal and external camera parameters, and from locations of the images of point in the films of the two cameras, 3D coordinates of the point can be calculated. Root mean square errors of the three coordinates of points in a large object volume (5m x 5m x 1.5m) were 15 mm, 13 mm, 13 mm and 6 mm, and relative errors in lengths averaged 0.5%, 0.7% and 0.5%, respectively.