Single Image Layer Separation Using Relative Smoothness

By Yu Li, Michael S. Brown. 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014

367 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/be0513d167e758635c83e14206de38f7becaee5d

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

Underexposed Photo Enhancement Using Deep Illumination Estimation (2019, background): “Smoothness Loss According to the smoothness prior [23, 20, 2], illumination in natural images is in general locally smooth.”
Cooperative Semantic Segmentation and Image Restoration in Adverse Environmental Conditions (2019, background): “Table 5 shows the comparison results with the stateof-the-art reflection removal methods, i,e., BDN [45], LB14[21], CEILNet [11], PLRS [52] and ERRNet [44].”
STAR: A Structure and Texture Aware Retinex Model (2019, methodology): “The Retinex model is in similar spirit with the intrinsic image decomposition model [1], [2], [35], which decomposes an observed image into Lambertian shading and reflectance (ignoring the specularity).”
Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks (2020, background): “It is reasonable to assume that the reflections (gray triangles) and the objects (blue circle) behind the glass have different distances to the camera [36], and the objects behind the glass is often focused (type (II) in Fig.”
Rain Streak Removal Using Layer Priors (2016, background): “More importantly, this inequality constraint plays a critical role in estimating reliable solutions as it regularizes the DC component of the recovered layers as verified in recent works by [15, 21].”
Intrinsic Decompositions for Image Editing (2017, neutral): “…gradients [Hor74], using a sparse norm on reﬂectance gradients [BST ∗ 14] or similarly, on differences between adjacent gradients [BHY15], using Gaussian Processes [CCFI14] or other probabilistic frameworks [LB14], or using a quadratic penaly term in a non-linear optimization energy [SYJL11,SY11].”
A Generic Deep Architecture for Single Image Reflection Removal and Image Smoothing (2017, methodology): “The regional gradients associated with these reflections can therefore be quite large, violating the assumption in [23].”
Structure-Revealing Low-Light Image Enhancement Via Robust Retinex Model (2018, background): “For simplicity, many works [16], [21], [22] often neglect the specular component C.”
A fusion-based enhancing method for weakly illuminated images (2016, methodology): “For example , Li and Brown [ 32 ] propose a fast separation method by building two likelihoods from the gradient histograms .”
Single Image Reflection Separation with Perceptual Losses (2018, methodology): “Recent researchers proposed methods for reflection removal from a single image [25, 21], but these approaches rely on hand-crafted priors such as ghost cues and relative smoothness which may not generalize to all images with reflection.”
LIME: Low-Light Image Enhancement via Illumination Map Estimation (2017, methodology): “The model of our problem is similar with that of the intrinsic image decomposition [16], [17], [18], [19], which attempts (b) Reflectancee”
Low-Light Image Enhancement via a Deep Hybrid Network (2019, background): “Retinex theory [34] assumes that the amount of light reaching observers can be decomposed into illumination and scene reflection [35]–[37].”