User Assisted Separation of Reflections from a Single Image Using a Sparsity Prior

By Anat Levin, Yair Weiss. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2004

433 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/ae4e96913be354bd512c242b9340b689923279ea

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

Single Image Layer Separation Using Relative Smoothness (2014, background): “One of the earliest work addressing intrinsic image decomposition was the Retinex algorithm [10] that employed simple heuristics that assumed strong edge gradients belonged to reflectance changes in the image.”
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].”
“Double-DIP”: Unsupervised Image Decomposition via Coupled Deep-Image-Priors (2018, background): “In transparency separation, the task is to separate the image into its superimposed reflection and transmission [37, 32, 26, 14].”
Single Image Reflection Separation with Perceptual Losses (2018, background): “The gradient sparsity prior is also explored together with optimal and minimum user assistance to better guide the illposed separation problem [17, 27].”
Blind deconvolution using a normalized sparsity measure (2011, neutral): “But since images are highly non-Gaussian [5], many approaches [6, 13, 14] instead use `p-norms on the gradients, with 0.”
Segmentation-Free Dynamic Scene Deblurring (2014, background): “To address this problem, many researchers studied the prior of latent image [14, 15, 18].”
Fast Image Deconvolution using Hyper-Laplacian Priors (2009, background): “Denoising [14], deblurring [3], transparenc y separation [11] and super-resolution [20], are all tasks that are inherently ill-posed.”
Blind Motion Deblurring Using Image Statistics (2006, background): “Recent work in computer vision applied natural images priors for a variety of applications like denoising [25, 24], super resolution [27], video matting [2], inpainting [16] and reflections decomposition [15].”
Using digital photography to study animal coloration (2007, background): “It is already possible to extract data of a scene from behind a plane of glass (Levin & Weiss, 2004), which could become useful for studies of aquatic organisms (although most glass filters out UV wavelengths; Lauziére et al., 1999).”
A Generic Deep Architecture for Single Image Reflection Removal and Image Smoothing (2017, methodology): “Obviously this is an ill-posed problem as there exist infinite feasible solutions, and hence most reflection removal algorithms require multiple input images [7, 30, 1, 19, 12, 22, 40, 41] or manual user interactions [21] to label reflectionand background-layer gradients, thus condensing the space o”
Efficient marginal likelihood optimization in blind deconvolution (2011, methodology): “Given k it solves a nonblind deconvolution problem, at which a mean latent image estimate μ is computed using iterative reweighted least squares [12, 13].”