彭亚丽, 刘芳, 焦李成, 刘侍刚. 基于秩4约束的遮挡点恢复方法[J]. 机器人, 2008, 30(2): 138-141,146.
引用本文: 彭亚丽, 刘芳, 焦李成, 刘侍刚. 基于秩4约束的遮挡点恢复方法[J]. 机器人, 2008, 30(2): 138-141,146.
PENG Ya-li, LIU Fang, JIAO Li-cheng, LIU Shi-gang. A Method for Occlusion Recovery Based on Rank 4[J]. ROBOT, 2008, 30(2): 138-141,146.
Citation: PENG Ya-li, LIU Fang, JIAO Li-cheng, LIU Shi-gang. A Method for Occlusion Recovery Based on Rank 4[J]. ROBOT, 2008, 30(2): 138-141,146.

基于秩4约束的遮挡点恢复方法

A Method for Occlusion Recovery Based on Rank 4

  • 摘要: 提出了一种基于秩4约束的遮挡点恢复方法.该方法首先假设所有的遮挡点及深度因子已知,利用所有遮挡点及深度因子组成一个秩4矩阵的特性,构造一个投影矩阵,并利用该投影矩阵求得所有的遮挡点及深度因子.然后将求得的遮挡点及深度因子代替图像中的遮挡点和深度因子,再求投影矩阵.经过多次迭代,最后求得遮挡点的真实图像位置.该方法的优点是平等地看待所有的图像及图像点.模拟实验和真实实验表明,该方法具有鲁棒性好、收敛性好及误差小等优点.

     

    Abstract: A method based on rank 4 is presented for occlusion recovery.Firstly,all the occlusions and depth factors are assumed known to obtain a rank 4 matrix that is used to construct a projective matrix.Then,all the occlusions and depth factors are obtained from the projective matrix.At the same time,the old occlusions and depth factors are substituted with the obtained ones to get a new projective matrix.After several iterations,the real values of the occlusions and depth factors can be found.The innovative point of the algorithm is that all the images and image points are treated uniformly.The experiments with both simulation and real data show that the method has such advantages as strong robustness,good convergence performance and small errors.

     

/

返回文章
返回