Blind Deblurring Using Internal Patch Recurrence
Tomer Michaeli and Michal Irani
Recurrence of small image patches across different scales of a natural
image has been previously used for solving ill-posed problems (e.g., superresolution
from a single image). In this paper we show how this multi-scale property
can also be used for "blind-deblurring", namely, removal of an unknown blur
from a blurry image. While patches repeat 'as is' across scales in a sharp natural
image, this cross-scale recurrence significantly diminishes in blurry images.
We exploit these deviations from ideal patch recurrence as a cue for recovering
the underlying (unknown) blur kernel. More specifically, we look for the blur
kernel k, such that if its effect is "undone" (if the blurry image is deconvolved
with k), the patch similarity across scales of the image will be maximized. We
report extensive experimental evaluations, which indicate that our approach compares
favorably to state-of-the-art blind deblurring methods, and in particular, is
more robust than them.
1. Paper [PDF]
2. Some visual comparisons [PDF]
3. Supplementary Material [PDF]
4. Matlab code [ZIP]
5. Database of 640 blurry images from [Sun et al., ICCP 2013] [ZIP] (link to the authors' website)
6. The deblurring results of 6 competing methods on the databse of [Sun et al., ICCP 2013] [ZIP] (link to the authors' website)
7. Our deblurring results on the databse of [Sun et al., ICCP 2013] [ZIP]
Note: errors are computed based on the best alignment to the ground-truth image, discarding 50 pixels from each border.