Acquiring linear subspaces for face recognition under variable lighting

By Kuang-chih Lee, Jeffrey Ho, David J. Kriegman. 2005

2,489 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

Privacy Risks of Securing Machine Learning Models against Adversarial Examples (2019, methodology): “The extended Yale Face database B is used to train face recognition models, and contains gray scale face images of 38 subjects under various lighting conditions [14, 30].”
Bayesian representation learning with oracle constraints (2015, methodology): “We use a relatively small dataset that is, however, well-suited to illustrate the features of the algo-rithm and facilitates the interpretation of the factorized latent spaces: the Yale Faces dataset (Lee et al., 2005).”
Half-Quadratic-Based Iterative Minimization for Robust Sparse Representation (2014, background): “Cropped and normalized 192 168 face images were captured under various controlled lighting conditions [49], [14].”
Supervised translation-invariant sparse coding (2010, result): “Tab. 4 shows our results compared with those reported in the [22] for Linear Discriminant Analysis (LDA)[3], Nearest Neighbor (NN), Nearest Subspace (NS)[ 15 ], and Sparse Representation (SR).”
Sparse Representation for Computer Vision and Pattern Recognition (2010, methodology): “2 left also compares to the Nearest Subspace method [45], which makes similar use of linear illumination models, but is not based on sparsity and does not correct sparse errors.”
Information Constraints on Auto-Encoding Variational Bayes (2018, positive): “Dataset The extended Yale B dataset [27] contains cropped faces [28] of 38 people under 50 lighting conditions.”
Latent Low-Rank Representation for subspace segmentation and feature extraction (2011, methodology): “extract salient features from corrupted data, using Extended Yale Database B [13], which consists of 2414 frontal face images of 38 classes.”
Robust Subspace Segmentation by Low-Rank Representation (2010, methodology): “We test LRR’s ability to cope with the large corruptions in data, using a part of Extended Yale Database B (Lee et al., 2005), which consists of 640 frontal face images of 10 classes (there are 38 classes in the whole database and we use the first 10 classes for experiments).”
A Weighted Variational Model for Simultaneous Reflectance and Illumination Estimation (2016, methodology): “In this experiment, the reflectance is estimated by the new model over four images [15, 36] under different illumination conditions.”
Disentangling and Learning Robust Representations with Natural Clustering (2019, methodology): “We preprocess the latter by only keeping images with illumination angles smaller than 100°and scaling images to 64×64, [16].”
Variational Tempering (2014, methodology): “The second data set contained 2000 face images (Yale Face Database B, cropped version2) from 28 individuals with 168×192 pixels in different poses and under different light conditions (Lee et al., 2005).”
Subspace Clustering by Block Diagonal Representation (2018, methodology): “We test on the Extended Yale B database [19].”
Manifold Relevance Determination (2012, methodology): “Yale faces: To show the ability of our method to model very high-dimensional spaces our ﬁrst experiment is applied to the Yale dataset (Georghiades et al., 2001; Lee et al., 2005) which contains images of several human faces under different poses and 64 illumination conditions.”
Gabor Feature Based Sparse Representation for Face Recognition with Gabor Occlusion Dictionary (2010, methodology): “We evaluated the performance of the proposed algorithm on three representative facial image databases: Extended Yale B [ 22 ], [18], AR [23] and FERET [24].”
Maximum Correntropy Criterion for Robust Face Recognition (2011, methodology): “Extended Yale B Database [49], [51]: The Extended Yale B database consists of 2,414 frontal face images from 38 subjects [49] under various lighting conditions.”
Linear Regression for Face Recognition (2010, methodology): “[21], [22], and AR [23] have been addressed.”