Single Image Deblurring Using Motion Density Functions

By Ankit Gupta, Neel Joshi, C. Lawrence Zitnick, Michael F. Cohen, Brian Curless. 2010

402 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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

Adversarial Spatio-Temporal Learning for Video Deblurring (2018, background): “Modern single-image deblurring methods iteratively estimate uniform or non-uniform blur kernels and the latent sharp image given a single blurry image [9]–[20].”
Fast removal of non-uniform camera shake (2011, methodology): “However, instead of applying the homographies to the sharp image f (as done in [24, 5]), we apply all possible homographies only once to a grid of single-pixel dots.”
Blind Image Deblurring Using Dark Channel Prior (2016, neutral): “movements) [8, 11, 28, 30, 33].”
Blind deconvolution using a normalized sparsity measure (2011, background): “[6] present a blur formation model based on in-plane rotations combined with translations.”
Deep Multi-scale Convolutional Neural Network for Dynamic Scene Deblurring (2016, background): “Most of these approaches are based on following blur model [29, 10, 13, 11]. where B , S and n are vectorizd blurry image, latent sharp image, and noise, respectively.”
Learning a convolutional neural network for non-uniform motion blur removal (2015, background): “In [7, 8, 25, 26, 31], nonuniform motion blur is modeled as a global camera motion, which basically estimates an uniform kernel in the camera motion space.”
Segmentation-Free Dynamic Scene Deblurring (2014, background): “Moreover, camera shake, including rotational movement, generalizes the 2D translational camera motion and it enables spatially varying blur by camera rotation [8, 10, 23].”
Human-Aware Motion Deblurring (2019, methodology): “Furthermore, rather than simply assuming the whole image is associated with a uniform blur kernel, some methods estimate a global blur kernel descriptor [45, 10], or predict a set of blur kernels [13, 11].”
Deblurring by Realistic Blurring (2020, background): “The blurring process can be mathematically formulated as [8, 38], where I B and I S are blurry and sharp images, respectively.”
A no-reference metric for evaluating the quality of motion deblurring (2013, result): “There has been work on estimating spatially-varying PSFs [Whyte et al. 2010; Gupta et al. 2010; Ji and Wang 2012]; however, as the dimensionality of the PSF increases, these methods are often less reliable and may produce more severe artifacts, as shown in a recent study [Köhler et al. 2012].”
DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks (2017, background): “[12] made an assumption that the blur is caused only by 3D camera movement.”
Online Video Deblurring via Dynamic Temporal Blending Network (2017, background): “To deal with spatially varying blur, often caused by rotational camera motion (roll) around the optical axis [34, 11, 14], additional non-trivial alignment of multiple images is required.”
$L_0$ -Regularized Intensity and Gradient Prior for Deblurring Text Images and Beyond (2017, methodology): “In [25], a similar model is proposed in which motion density functions are used to represent camera motion trajectories.”
Learning to Estimate and Remove Non-uniform Image Blur (2013, neutral): “In summary, previous approaches to our deblurring problem either (i) fall short in the estimation or the deblurring step, (ii) require multiple images, or (iii) consider a limited set of blur types (e.g., uniform, constant velocity motion, camera rotation).”
Non-uniform Deblurring for Shaken Images (2010, background): “Gupta et al. (2010) propose a model which is similar in spirit to our own, recovering a set of weights over a 3D parameter space which describes transformations of the sharp image.”