Stochastic Deconvolution

Gregson, James; Heide, Felix; Hullin, Matthias B.; Rouf, Mushfiqur; Heidrich, Wolfgang

doi:10.1109/cvpr.2013.139

Cited by 11 publications

(19 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In earlier work, we presented stochastic random-walk optimization for tomography [38] and non-blind deblurring [3] that uses many incremental local solution updates at sampled locations. Sample placement is driven by a stochastic random walk that favors local exploration where fast progress has recently been made.…”

Section: Related Workmentioning

confidence: 99%

“…Given the kernel estimated at the current scaleK s , the function updateIntrinsicImage() in Algorithm 1 updates our estimate of the intrinsic image by solving a non-blind deconvolution problem using a stochastic random walk optimization (Algorithm 2, also see [3]), which minimizes objectives of the form: …”

Section: A Updating the Intrinsic Imageî Smentioning

confidence: 99%

“…Following the analysis in [3], our stochastic random walk framework is a form of stochastic coordinate descent (SCD, [41], [42], [43]). In each iteration, the algorithm picks a single pixel in the image and checks if the objective can be reduced by depositing energy in this pixel.…”

Section: ) Empirical Convergencementioning

confidence: 99%

“…In this paper we extend our recent work in stochastic non-blind deconvolution [3] to derive a blind method that is purely based on stochastic sampling. The method relies entirely on local evaluations of the objective function, without the need to compute gradients.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Stochastic Blind Motion Deblurring

Xiao

Gregson

Heide

et al. 2015

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

Blind motion deblurring from a single image is a highly under-constrained problem with many degenerate solutions. A good approximation of the intrinsic image can therefore only be obtained with the help of prior information in the form of (often non-convex) regularization terms for both the intrinsic image and the kernel. While the best choice of image priors is still a topic of ongoing investigation, this research is made more complicated by the fact that historically each new prior requires the development of a custom optimization method. In this paper, we develop a stochastic optimization method for blind deconvolution. Since this stochastic solver does not require the explicit computation of the gradient of the objective function and uses only efficient local evaluation of the objective, new priors can be implemented and tested very quickly. We demonstrate that this framework, in combination with different image priors produces results with PSNR values that match or exceed the results obtained by much more complex state-of-the-art blind motion deblurring algorithms.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Updating the Intrinsic Imageî Smentioning

confidence: 99%

Section: ) Empirical Convergencementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Stochastic Blind Motion Deblurring

Xiao

Gregson

Heide

et al. 2015

IEEE Trans. on Image Process.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Hence, numerous deconvolution algorithms have been developed to estimate the latent sharp image. They include Wiener [2] and Richardson-Lucy (RL) [3,4] techniques, least squares minimization [5], Bayesian inference [6][7][8], advanced variational based [9][10][11][12][13], and stochastic framework [14] methods. A challenging problem in latent image restoration is the presence of wavelike artifacts called ringing that appear near strong edges.…”

Section: (C) Blur Kernel (Psf) (D)-(g)mentioning

confidence: 99%

Image Deconvolution Ringing Artifact Detection and Removal via PSF Frequency Analysis

Mosleh

Langlois

Green³

2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. We present a new method to detect and remove ringing artifacts produced by the deconvolution process in image deblurring techniques. The method takes into account non-invertible frequency components of the blur kernel used in the deconvolution. Efficient Gabor wavelets are produced for each non-invertible frequency and applied on the deblurred image to generate a set of filter responses that reveal existing ringing artifacts. The set of Gabor filters is then employed in a regularization scheme to remove the corresponding artifacts from the deblurred image. The regularization scheme minimizes the responses of the reconstructed image to these Gabor filters through an alternating algorithm in order to suppress the artifacts. As a result of these steps we are able to significantly enhance the quality of the deblurred images produced by deconvolution algorithms. Our numerical evaluations using a ringing artifact metric indicate the effectiveness of the proposed deringing method.

show abstract