光流场重建线性算法的实验模拟

Experiment simulation of 3D reconstruction linear algorithm based on optical flow

由光流场重建三维运动和结构是计算机视觉研究中的热点和前沿 ,它分为线性算法和非线性算法。非线性算法的初始值选取非常重要 ,否则会导致算法失败。本文提出了一种由光流场重建三维结构线性算法的实验模拟新方法 ,该方法巧妙地利用合成数据和图像序列验证了重建的线性算法 ,该思路也适用于其它算法。

Reconstructing 3D motion and structure based on optical flow becomes the hot point and frontier in the research of computer vision. It is divided into linear and unlinear algorithms. The choice of initial value of the unlinear algorithm is important, otherwise algorithm fails. We present a new experiment simulation method to prove linear algorithm, which determines 3D motion and structure based on optical flow. The presented Experimental method using synthetic data and image sequences provided a good estimation of the optical flow and 3D motion and structure.