Centre for Artificial Intelligence and Robotics, Bangalore.
IEEE Trans Image Process. 1997;6(9):1246-54. doi: 10.1109/83.623188.
The issues governing the computation of optical flow in image sequences are addressed. The trade-off between accuracy versus computation cost is shown to be dependent on the redundancy of the image representation. This dependency is highlighted by reformulating Horn's (1986) algorithm, making explicit use of the approximations to the continuous basis functions underlying the discrete representation. The computation cost of estimating optical flow, for a fixed error tolerance, is shown to be a minimum for images resampled at twice the Nyquist rate. The issues of derivative calculation and multiresolution representation are also briefly discussed in terms of basis functions and information encoding. A multiresolution basis function formulation of Horn's algorithm is shown to lead to large improvements in dealing with high frequencies and large displacements.
本文讨论了在图像序列中计算光流的问题。结果表明，精度与计算成本之间的权衡取决于图像表示的冗余度。通过重新表述 Horn 的（1986 年）算法并明确利用离散表示所基于的连续基函数的近似值，突出了这种依赖性。对于固定的误差容限，以奈奎斯特速率的两倍重采样的图像的光流估计计算成本最小。还简要讨论了基于基函数和信息编码的导数计算和多分辨率表示的问题。表明 Horn 算法的多分辨率基函数公式在处理高频和大位移方面有很大的改进。
