Super-resolution Using Sub-space Models
In the previous chapter we demonstrated that generic MRF prior models can go a long way towards improving the performance of our super-resolution estimators. However, the generality of these priors is also their weakness, and in computing a superresolution MAP estimate, there is a trade-off to be made between reduced noise and excessive smoothness.
KeywordsInput Image Face Image Training Image Reconstruction Error Real Image
Unable to display preview. Download preview PDF.