Toward a Practical Face Recognition System: Robust Alignment and Illumination by Sparse Representation

By Andrew Wagner, John Wright, Arvind Ganesh, Zihan Zhou, Hossein Mobahi, Yi Ma. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012

680 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/25fce91ce1b974865506c14d2e4714d8db2672d1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

In Computer Vision and Pattern Recognition

Citations in the corpus

Deep Learning Identity-Preserving Face Space (2013, background): “[26] designed a projector-based system to capture images of each subject in the gallery under a few illuminations, which can be linearly combined to generate images under arbitrary illuminations.”
Discriminant Correlation Analysis: Real-Time Feature Level Fusion for Multimodal Biometric Recognition (2016, background): “Most recently, sparse representation has attracted the interest of many researchers, both for reconstructive and discriminative tasks [18]–[20].”
Half-Quadratic-Based Iterative Minimization for Robust Sparse Representation (2014, methodology): “In addition, soft-thresholding function and iteratively reweighted methods have drawn much attention in subspace segmentation [11], robust alignment [58], nuclear norm minimization [59], [60], and structure sparsity [61], [62] where the used minimization functions are related to HQ optimization.”
Inertial Sensor-Based Gait Recognition: A Review (2015, background): “…gait recognition in a global manner and side-by-side with more conventional recognition systems, it should be mentioned that reported performance metrics are fully comparable i.e. , with the results of the latest state-of-the-art in the ﬁeld of face recognition in realistic circumstances [84,85].”
Sparse Representation Based Fisher Discrimination Dictionary Learning for Image Classification (2014, methodology): “…et al. 2009; Ramirez et al. 2010; Yang et al. 2010b; Rodriguez and Sapiro 2007; Mairal et al. 2012), natural image classification (Rodriguez and Sapiro 2007; Yang et al. 2009), and human action recognition (Qiu et al. 2011; Guha and Ward 2012; Wang et al. 2012; Castrodad and Sapiro 2012), etc.”
Toward Guaranteed Illumination Models for Non-convex Objects (2013, methodology): “These models have been used for recognition in many subsequent works [9, 20, 21, 19], and have been extended in a number of directions [8, 16].”
Sparse Representation for Computer Vision and Pattern Recognition (2010, background): “However, under mild conditions [28], the desired solution (α 0 , e 0 ) is 2 For a detailed explanation of how such images can be obtained, see [68] .”
Partial Face Recognition: Alignment-Free Approach (2013, methodology): “3) We extend the MKD-SRC method for partial face verification, which is effective on the LFW database.”
Extended SRC: Undersampled Face Recognition via Intraclass Variant Dictionary (2012, background): “3 percent on the dup2 set [13].”
Large-Pose Face Alignment via CNN-Based Dense 3D Model Fitting (2016, background): “…such as eye corners, nose tip, etc. Improving the face alignment accuracy is beneficial for many computer vision tasks related to facial analysis, because it is used as a prerequisite for these tasks, e.g., face recognition [28], 3D face reconstruction [20, 21] and face de-identification [10].”
The generalized lasso is reducible to a subspace constrained lasso (2013, background): “Indeed, such representations have proven effective in applications ranging from image restoration [2, 3], face recognition [4, 5], object recognition [6], speech classification [7], speech recognition [8], music genre classification [9], and topic detection in text documents [10].”
Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network (2018, methodology): “In the beginning, face alignment that aims at detecting a special 2D ﬁducial points [1,2,3,4] is commonly used as a prerequisite for other facial tasks such as face recognition [5] and assists 3D face reconstruction [6,7] to a great extent.”
Deep Networks from the Principle of Rate Reduction (2020, background): “8Notice that similar sparse representation models have long been proposed and used for classification purposes in applications such a face recognition, demonstrating excellent effectiveness (Wright et al., 2009; Wagner et al., 2012).”
Sparse Modeling for Image and Vision Processing (2014, background): “To deal with that issue, Wagner et al. [2012] have proposed dedicated solutions, which jointly optimize the variables α and e, along with a deformation variable whose purpose is to automatically “realign images”.”