Similarity Scores Based on Background Samples

By Lior Wolf, Tal Hassner, Yaniv Taigman. Asian Conference on Computer Vision, 2009

313 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/82f3b7cacc15e026fd3a7639091d54162f6ae064

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

Effective Unconstrained Face Recognition by Combining Multiple Descriptors and Learned Background Statistics (2011, unknown): “Since its recent publication, a lot of attention has been focused on improving performance on the benchmarks associated with the LFW database (see, e.g., [1], [2], [3], [6], [14], [15], [16], [17]).”
Face recognition in unconstrained videos with matched background similarity (2011, background): “It is thus related to a recent family of similarities [15, 22, 23] based on a background set of examples and which employ classifiers.”
Describable Visual Attributes for Face Verification and Image Search (2011, methodology): “Neither do many other recent methods on LFW [48], [56], [60], [61], all of which use a large set of carefully designed local features.”
Learning Discriminant Face Descriptor (2014, methodology): “In this experiment, we use the aligned images (LFW-a) [48] and crop the images with the size of 150 (cid:4) 130 from the original images.”
Tom-vs-Pete Classifiers and Identity-Preserving Alignment for Face Verification (2012, methodology): “It is well known that alignment is critical for good performance in face recognition with uncontrolled images [11, 31, 33].”
Learning hierarchical representations for face verification with convolutional deep belief networks (2012, methodology): “For our experiments, we used the LFW-a(2) face images aligned using a commercial face alignment software, provided in [38].”
Distance Metric Learning with Eigenvalue Optimization (2012, methodology): “Recently it has become a benchmark to test new face verification algorithms (Wolf et al., 2008; Guillaumin et al., 2009; Wolf et al., 2009; Taigman et al., 2009; Pinto et al., 2011).”
Labeled Faces in the Wild: A Survey (2016, methodology): “, 2011 [106] (Journal verson of [96] and [105]), Cosine, 2011 [72], Large scale feature search, 2011 [31], LARK, 2011 [84], DML-eigen, 2012 [111], Conv-DBN, 2012 [48], CMD (Ensemble metric), 2012 [43], LBP PLDA, 2012 [62], MRF-MLBP, 2013 [9], Pose-robust, 2013 [108], Similarity metric, 2013 [23], DF”
Discriminative Deep Metric Learning for Face Verification in the Wild (2014, methodology): “As suggested in [16, 26, 36], we also use the square root of each feature and evaluate the performance of our DDML method when all the six different feature descriptors are combined.”
Learning to Align from Scratch (2012, background): “for algorithms that were explicitly designed to be robust to some misalignment [1].”
Face recognition with learning-based descriptor (2010, methodology): “Currently, descriptor-based approaches [10, 20, 31] have been proven to be effective face representations producing best performance [8, 18, 12].”
Learning Locally-Adaptive Decision Functions for Person Verification (2013, methodology): “MERL+Nowak [14] 76.2 LDML [12] 79.3 LBP + CSML [19] 85.6 CSML + SVM [19] 88.0 Combined b/g samples [25] 86.8 DML-eig combined [27] 85.7 Deep Learning combined [15] 87.8 Our method 89.6 the best reported results 3 in LFW under the category of “no outside data is used beyond alignment/feature…”
One Shot Similarity Metric Learning for Action Recognition (2011, result): “OSS results consequently outperformed other methods on the LFW challenge [20].”
Effective face frontalization in unconstrained images (2014, background): “…that pool information over large image regions, thereby accounting for possible misalignments due to pose changes (e.g., [20, 25, 33, 36]), by improving 2D face alignment accuracy [14, 15, 35], or by using massive face collections to learn pose-robust representations [13, 26, 27, 41].”
Cosine Similarity Metric Learning for Face Verification (2010, background): “Each type of feature is tested with the original feature vector or the square root of the feature vector [8,14,20].”
Fisher Vector Faces in the Wild (2013, background): “Another popular approach is based on exemplar SVMs [32] [37] [38].”
Rectified Linear Units Improve Restricted Boltzmann Machines (2010, unknown): “0034 (Wolf et al., 2009), and 0 .”
POOF: Part-Based One-vs.-One Features for Fine-Grained Categorization, Face Verification, and Attribute Estimation (2013, background): “Face recognition is an extreme case of fine-grained visual categorization in which the “subcategories” are individual instances, and the best face recognition methods extract features from locations determined by finding facial landmarks such as the corners of the eyes [3, 31, 36].”
Similarity Metric Learning for Face Recognition (2013, methodology): “Recently, considerable research efforts are devoted to the unconstrained face verification problem [8, 17, 18, 20, 23, 24], the task of which is to predict whether two face images represent the same person or not.”