Entropy Minimization for Shadow Removal

By Graham David Finlayson, Mark Samuel Drew, Cheng Lu. International Journal of Computer Vision, 2009

381 citations recorded by the source corpus.

Subfields: Atomic and Molecular Physics, and Optics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/cfc20a3143a9267c7b29dbe3965cc6036d0504e8

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

Citations in the corpus

A Multi-Task Mean Teacher for Semi-Supervised Shadow Detection (2020, methodology): “Existing methods detect shadows by developing physical models of color and illumination [6, 5], by using data-driven input images ground truths our method DSD [41] CVPR’2019 BDRAR [43] ECCV’2018 Figure 1: Shadow detection on two inputs with a soft shadow (the ﬁrst row) and multiple shadow regions…”
Ieee Transactions on Pattern Analysis and Machine Intelligence Shape, Illumination, and Reflectance from Shading (neutral): “We are not the first to explore global parsimony priors on reflectance: different forms of this idea have been used in intrinsic images techniques [15], photometric stereo [46], shadow removal [47], and color representation [48].”
CrackTree: Automatic crack detection from pavement images (2012, methodology): “Compared to many classical shadow-removal algorithms (Finlayson et al., 2006, 2009; Arbel and Hel-Or, 2007), the geodesic shadow-removal algorithm 0167-8655/$ -see front matter Ó 2011 Elsevier B.V.”
DeshadowNet: A Multi-context Embedding Deep Network for Shadow Removal (2017, neutral): “Conventional methods for shadow detection either lack robust shadow features [35, 20, 15], or can only be applied to high-quality images [8, 30].”
Portrait shadow manipulation (2020, neutral): “Shadow removal has also been formulated as an entropy minimization problem (Finlayson et al. 2009, 2002), where invariant chromaticity and intensity images are used to produce a shadow mask that is then re-integrated to form a shadow-free image.”
Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal (2017, background): “[5, 7] provide the illumination invariant solutions that work well only on high quality images.”
Single-image shadow detection and removal using paired regions (2011, background): “Many approaches are motivated by physical models of illumination and color [12, 15, 16, 7, 5].”
Shape, albedo, and illumination from a single image of an unknown object (2012, methodology): “We use the entropy formulation of [22], which has been previously used for the related task of shadow removal [9], to impose a similar prior on global albedo entropy.”
Color Constancy, Intrinsic Images, and Shape Estimation (2012, methodology): “R´enyi measures of entropy are quadratically expensive to compute naively, so others have used the Fast Gauss Transform [2] and histogram-based techniques [1] to approximate them in linear time.”