Image and depth from a conventional camera with a coded aperture

By Anat Levin, Rob Fergus, Frédo Durand, William T. Freeman. 2007

1,566 citations recorded by the source corpus.

Subfields: Atomic and Molecular Physics, and Optics, 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 Atomic and Molecular Physics, and Optics

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Depth Map Prediction from a Single Image using a Multi-Scale Deep Network (2014, methodology): “[10] perform depth from defocus using a modified camera aperture, while the Kinect and Kinect v2 use active stereo and time-of-flight to capture depth.”
Blind Deblurring Using Internal Patch Recurrence (2014, methodology): “However, their error ratios were computed with the non-blind deblurring of [11].”
Blind deconvolution using a normalized sparsity measure (2011, methodology): “We solve the inner IRLS system to a low level of accuracy, using a few (less than 5) conjugate gradient (CG) iterations.”
Two-Phase Kernel Estimation for Robust Motion Deblurring (2010, methodology): “(a) Noisy input (b) [15] (c) [17] (d) Ours”
A no-reference metric for evaluating the quality of motion deblurring (2013, background): “• Sparsity priors (Sps) [Levin et al. 2007] assume that gradient magnitudes of natural images follow a heavy-tailed distribution.”
End-to-end optimization of optics and image processing for achromatic extended depth of field and super-resolution imaging (2018, methodology): “…used for tasks as diverse as motion deblurring [Raskar et al. 2006], defocus deblurring [Zhou et al. 2009; Zhou and Nayar 2009], depth estimation [Levin et al. 2007, 2009], multispectral imaging [Wagadarikar et al. 2008], light field imaging [Marwah et al. 2013; Ng et al. 2005; Veeraraghavan et…”
Accurate Blur Models vs. Image Priors in Single Image Super-resolution (2013, methodology): “The sparse prior was optimized using the iterative re-weighted least squares algorithm [14], with parameters chosen using a separate training set.”
On Bayesian Adaptive Video Super Resolution (2014, methodology): “We used a state-of-the-art deconvolution method [17] to postprocess its output.”
Deep Convolutional Neural Network for Image Deconvolution (2014, methodology): “The sparse image priors follow heavy-tailed distributions, such as a Gaussian Mixture Model [1, 11] or a hyper-Laplacian [7, 3], which could be efficiently optimized using half-quadratic (HQ) splitting [3].”
A Generalized Iterated Shrinkage Algorithm for Non-convex Sparse Coding (2013, methodology): “Several typical algorithms include iteratively reweighted least squares (IRLS) [12, 14, 23, 24, 28], iteratively reweighted 1 -minimization (IRL1) [8], iteratively thresholding method (ITM-(cid:2) p ) [33, 34], and look-up table (LUT) [25].”
Fast Image Deconvolution using Hyper-Laplacian Priors (2009, methodology): “Our LUT algorithm is more than100 times faster than the IRLS method of [10].”
Image deblurring and denoising using color priors (2009, methodology): “While such a prior does disambiguate the solution, it can result in an overly-smooth solution and introduce ringing artifacts [12].”
Analyzing spatially-varying blur (2010, methodology): “For example, for defocus blur where the kernel changes with scene depth, one can imagine using the framework in [16] but doing away with the requirement for special capture conditions by using the proposed likelihood measure in (12) instead of cues from a coded aperture.”
Image Restoration Using Convolutional Auto-encoders with Symmetric Skip Connections (2016, methodology): “kernel tpye Krishnan [57] Levin [58] Cho [59] Schuler [60] Xu [56] RED30”