A fast super-resolution reconstruction algorithm for pure translational motion and common space-invariant blur

By Michael Elad, Yacov Hel-Or. 21st IEEE Convention of the Electrical and Electronic Engineers in Israel. Proceedings (Cat. No.00EX377), 2000

528 citations recorded by the source corpus.

Super-resolution reconstruction deals with the recovery of a single high-resolution image from a set of warped, blurred and decimated versions of it. Several algorithms were already proposed for the solution of this general problem. We concentrate on a special case where the warps are pure translations, the blur is space invariant and the same for all the images, and the noise is white. We exploit previous results to develop a new highly efficient super-resolution reconstruction algorithm for this case, which separates the treatment to de-blurring and measurements fusion. The fusion part is shown to be a non-iterative and very simple algorithm, preserving the optimality of the entire reconstruction process, in the maximum-likelihood sense. Simulations demonstrate the capabilities of the proposed algorithm.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/00d3e9b9161be40d7e9adba9cb6110df5f1e4d81

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Advances and challenges in super‐resolution (2004, background): “If the optical blur of the imaging system is translation invariant, then the order of the operations of the image shift and image blur are commutative (Elad and Hel-Or, 2001).”
Image super-resolution: The techniques, applications, and future (2016, background): “In general, if there is supplementary information among the images, SR is feasible [22].”
Super-resolution image reconstruction: a technical overview (2003, background): “Elad and Hel-Or [50] proposed an SR algorithm that separates fusion and deblurring.”
Generalizing the Nonlocal-Means to Super-Resolution Reconstruction (2009, background): “The idea of breaking the super-resolution task into two parts—fusing the inputs and then deblurring—has been suggested previously in [13], [18], [20], and [21].”
Fast and robust multiframe super resolution (2004, methodology): “Fig. 2(d) shows the effect of this error in the motion model (shadows around Lena’s hat) when the non robust least-squares approach [ 5 ] is used for reconstruction.”
Super-Resolution Without Explicit Subpixel Motion Estimation (2009, background): “In particular, in order to be effective, all the existing super-resolution approaches must perform (sub-pixel) accurate motion estimation [1], [2], [3], [4], [5], [6], [7], [8], [9], [10].”
Detail-Revealing Deep Video Super-Resolution (2017, unknown): “With certain assumptions [5, 7], W 0→iS SW0→i becomes a diagonal matrix.”
Imaging without lenses: achievements and remaining challenges of wide-field on-chip microscopy (2012, background): “The first of these computational blocks, termed ‘pixel super-resolution’ 45–47 , is used to overcome the resolution limitation due to the pixel size and is required to achieve subpixel spatial resolution in lens-free on-chip imaging 6–10 .”
Robust Super-resolution Volume Reconstruction from Slice Acquisitions: Application to Fetal Brain MRI (2010, methodology): “The solution to (6) is equivalent to the pseudo-inverse and least-squares solution of (1), and is statistically equal to the computation of the sample mean [20] if all the error samples have the same variance.”