Removing Camera Shake from a Single Photograph

By Sundaresh Ram, S. Jayendran. 2009

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

Verifying Neural Networks with Mixed Integer Programming (2017, neutral): “These blurring perturbations are meant to mimic blur from camera shake [7].”
Examining the Impact of Blur on Recognition by Convolutional Networks (2016, background): “Graphics and vision researchers have made significant progress on this problem across the last decade [7, 13, 15, 28]—including with recent neural network-based methods [1, 21, 22, 26].”
Learning to Deblur (2014, unknown): “, [1], [2], [3].”
Conditioned Regression Models for Non-blind Single Image Super-Resolution (2015, background): “, [9, 18, 19]) are typically slow and take several minutes per image, making them impractical to build a large corpus of training data.”
Single Image Layer Separation Using Relative Smoothness (2014, background): “Our work is inspired by the success of imposing a gradient sparsity constraint on the images, however, in our problem, the two layers’ gradients do not have the same distributions.”
What makes a good model of natural images? (2007, background): “This includes problems such as transparency analysis [13], removal of camera blur [6] image denoising and image inpainting [19].”
On Bayesian Adaptive Video Super Resolution (2014, methodology): “Future work will incorporate the recent developments in each subproblem, such as highorder image prior model [29], nonlocal motion prior model [35], feature matching for fast moving objects [7], [33], [40], and advanced inference methods for estimating the spatially variant blur kernel [9], [39].”
Ieee Transactions on Pattern Analysis and Machine Intelligence Shape, Illumination, and Reflectance from Shading (unknown): “To construct these priors we build upon past work on natural image statistics, which has demonstrated that simple statistics govern local patches of natural images [2], [3], [37], and that these statistics can be used for denoising [38], inpainting [39], deblurring [40], etc.”
Non-uniform Deblurring for Shaken Images (2010, methodology): “To approach these two problems, we apply our model within the frameworks proposed by Miskin & MacKay [16] and Fergus et al. [10] for the blind case, and Yuan et al. [29] for the case of a noisy / blurred image pair.”
Deep Mean-Shift Priors for Image Restoration (2017, background): “In the last decades, several natural image priors have been proposed, including total variation [29], gradient sparsity priors [12], models based on image patches [5], and Gaussian mixtures of local filters [25], just to name a few of the most successful ideas.”
Image super-resolution using gradient profile prior (2008, unknown): “This kind of “sparseness prior” has been successfully applied to super-resolution[28], denoising [23] [24], inpainting [17], transparency separation [16] and deblurring [8, 15].”
Image Completion Approaches Using the Statistics of Similar Patches (2014, methodology): “Gradient-domain statistics have been applied in denoising [44], deconvolution [45], and diffusion-based inpainting [3].”
Nonparametric Blind Super-resolution (2013, result): “This is in contrast to the MAPk method [5, 13, 14] which seeks to estimate k alone, by maximizing its marginal posterior, p(k|l).”
Single Image Deblurring Using Motion Density Functions (2010, methodology): “[4] recover a blur kernel by using a natural image prior on image gradients in a variational Bayes framework.”
From Motion Blur to Motion Flow: A Deep Learning Solution for Removing Heterogeneous Motion Blur (2016, methodology): “nificant progress in image deblurring [9, 6, 42, 27, 28, 10], most work focuses on spatially-uniform blur.”
Blind Deblurring Using Internal Patch Recurrence (2014, background): “Some studies assume simple parametric probability models, which promote sparsity of image gradients [5,17,12,13,10].”