How well can wavelet denoising improve the accuracy of computing fundamental matrices?

The existence of noise is a serious obstacle for solving many computer vision problems. Computing the fundamental matrix is a typical one. Motivated by the success of the wavelet denoising technique in image processing, we study the interesting topic of how well wavelet denoising can improve the accuracy of computing fundamental matrices. The answer depends on: 1) what wavelets should be applied; 2) what kind of images wavelets could be applied to in order to improve the accuracy of their fundamental matrix. Experimental results show that wavelet denoising is promising in computer vision as well as in image processing.