Analysis of Magnification in Depth from Defocus
This work addresses accuracy issues in DFD for computer vision applications, but it is incremental as it focuses on analyzing existing methods rather than introducing new ones.
The paper tackles the problem of magnification effects in depth from defocus (DFD) by analyzing how image warping compares to optical methods, finding that ignoring scale factors introduces bias and reduces efficiency in blur estimation.
In depth from defocus (DFD), when images are captured with different camera parameters, a relative magnification is induced between them. Image warping is a simpler solution to account for magnification than seemingly more accurate optical approaches. This work is an investigation into the effects of magnification on the accuracy of DFD. We comment on issues regarding scaling effect on relative blur computation. We statistically analyze accountability of scale factor, commenting on the bias and efficiency of the estimator that does not consider scale. We also discuss the effect of interpolation errors on blur estimation in a warping based solution to handle magnification and carry out experimental analysis to comment on the blur estimation accuracy.