Blessing of Dimensionality: High-Dimensional Feature and Its Efficient Compression for Face Verification

By Dong Chen, Xudong Cao, Fang Wen. 2013

697 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

Faceless Person Recognition: Privacy Implications in Social Media (2016, background): “Like other areas of computer vision, the last years have seen a shift from classifiers based on handcrafted features and metric learning approaches [13, 14, 15, 16, 17, 18, 19] towards methods based on deep learning [20, 21, 22, 23, 24, 25, 26, 27].”
Deep Learning Face Representation by Joint Identification-Verification (2014, methodology): “Joint Bayesian has been successfully used to model the joint probability of two faces being the same or different persons [3, 4].”
Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis (2017, methodology): “One category tries to adopt hand-crafted or learned pose-invariant features [4,28], while the other resorts to synthesis techniques to recover a frontal view image from a large pose face image and then use the recovered face images for face recognition [45,46].”
Naive-Deep Face Recognition: Touching the Limit of LFW Benchmark or Not? (2015, background): “Extensive works have been done to push the accuracy limit on it [3, 16, 4, 1, 2, 5, 11, 10, 12, 14, 13, 17, 9].”
Web-scale training for face identification (2014, background): “This might be counter intuitive, since the common wisdom [3] is that the representation should be as large as the number of samples can support.”
Surpassing Human-Level Face Verification Performance on LFW with GaussianFace (2014, methodology): “Although there has been significant work [22, 9, 5, 14, 47, 13, 59, 50, 51, 53] on LFW and the accuracy rate has been improved from 60.”
Person Recognition in Personal Photo Collections (2015, background): “Both settings have seen a recent shift from sophisticated classifiers based on hand-crafted features and metric learning approaches [20, 7, 5, 29, 26, 41, 1], towards methods based on deep learning [37, 36, 43, 33, 27, 38, 21].”
Multi-View Perceptron: a Deep Model for Learning Face Identity and View Representations (2014, methodology): “The best practise of face recognition extracts the above features on the landmarks of face images with multiple scales and concatenates them into high dimensional feature vectors [4, 21].”
DeepFace: Closing the Gap to Human-Level Performance in Face Verification (2014, background): “Moreover, the system produces an extremely compact face representation, in sheer contrast to the shift toward tens of thousands of appearance features in other recent systems [5, 7, 2].”
Cross-Age Reference Coding for Age-Invariant Face Recognition and Retrieval (2014, methodology): “We compare CARC to several different state-of-the-art methods, including: (1) high-dimensional local binary pattern [4] (HD-LBP), the local feature used for cross-age celebrity coding.”
A Practical Transfer Learning Algorithm for Face Verification (2013, methodology): “This fact is commonly approximated by x = µ + , (1) where x is the assumed facial feature, e.g., LBP, SIFT, etc., which is linearly decomposed into two independent variables related to identity µ and intra-personal variations .”
Marginal Loss for Deep Face Recognition (2017, result): “As can be seen from Table 2, the proposed marginal loss significantly outperforms all the published results [4, 6, 3, 26] on this dataset, even surpassing the human-level performance with a clear margin.”
Learning Deep Face Representation (2014, background): “Pete+Attribute [18] LBP [2] Pyramid CNN (ours) Human on cropped faces”