In search of illumination invariants

By Hansen F. Chen, Peter N. Belhumeur, David W. Jacobs. Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662), 2001

363 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/9cab8e79ee34b346fae514310ad0402cf5e3119c

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

Attribute and simile classifiers for face verification (2009, background): “Our low-level features are designed following a great deal of work in face recognition (and the larger recognition community) which has identified gradient direction and local descriptors around fiducial features as effective first steps toward dealing with illumination [6, 28, 22, 23, 10].”
Shedding light on stereoscopic segmentation (2004, methodology): “The effect of changing lighting on the object appearance (for fixed viewpoint) was analyzed by Belhumeur et al. [26] [3].”
Removing photography artifacts using gradient projection and flash-exposure sampling (2005, methodology): “Gradient-based invariants have been used for shadow removal [Finlayson et al. 2002] and face recognition [Chen et al. 2000].”
From Few to Many: Illumination Cone Models for Face Recognition under Variable Lighting and Pose (2001, result): “It is, in spirit, most closely related to the synthesis approaches suggested in [61, 56] and stands in stark contrast to the illumination insensitivity techniques argued for in [3, 8].”
Visual Tracking via Locality Sensitive Histograms (2013, methodology): “To deal with this problem, numerous methods have been proposed based on illumination invariant features [5].”
Clustering appearances of objects under varying illumination conditions (2003, background): “The important conclusion of [5] is that while the lighting direction can be random, the direction of the image gradient is not.”
Shape Classification Using the Inner-Distance (2007, methodology): “Instead of color information, we use gradient orientation, which is more robust to lighting change [10], which is very important for classification tasks.”
Bayesian models of object perception. (2003, background): “They constrain local boundary detection (Figure 3a; [21 ]), identify regularities on the changes of image features as illumination conditions vary (Figure 3b; [22]), and help to interpret human body motion [23].”
On the set of images modulo viewpoint and contrast changes (2009, background): “For the case of illumination, it has been shown [5] that general-case (global) illumination invariants do not exist, even for Lambertian objects.”
Bayesian Defogging (2012, background): “…albedo · Scene depth · Markov random field · Bayesian estimation K. Nishino ( ) · L. Kratz · S. Lombardi Department of Computer Science, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104, USA e-mail: kon@drexel.edu L. Kratz e-mail: lak24@drexel.edu S. Lombardi e-mail: sal64@drexel.edu”
Enhanced Local Texture Feature Sets for Face Recognition Under Difficult Lighting Conditions (2007, neutral): “Output Raw images”
Face Recognition Under Varying Illumination Using Gradientfaces (2009, background): “However, the authors in [24] show that there are no illumination invariants, the illumination invariant obtained by the above-mentioned methods is at best insensitive to illumination.”