Patent
Face authentication method and device
Ning Wang,Xiangde Zhang,Wuming Jiang,Yaoluo Zhang +3 more
- 17 Oct 2014
77
TL;DR: In this article, a face authentication method was proposed, which includes: acquiring multiple face training images, extracting Gabor features of the multiple faces, extracting Pattern of Oriented Edge Magnitude (POEM) features of multiple face images, fusing the Gabor and POEM features, and training the positive samples and negative samples of multiple faces using an AdaBoost algorithm.
read more
Abstract: The disclosure discloses a face authentication method and device. The face authentication method includes: acquiring multiple face training images; extracting Gabor features of the multiple face training images; extracting Pattern of Oriented Edge Magnitude (POEM) features of the multiple face training images; fusing the Gabor features of the multiple face training image and the POEM features of the multiple face training image to acquire positive samples and negative samples of the multiple face training images; training the positive samples and negative samples of the multiple face training images to obtain training results by an AdaBoost algorithm; and performing face authentication by the training results. By the disclosure, the problem of difficulty of a face authentication method in the related technology in combination of efficiency and recognition rate is solved, and the effects of improving feature extraction efficiency of face recognition and increasing the face recognition rate are further achieved.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Image identification method and device
Yang Song,Chen Zhijun,Wan Shaohua +2 more
- 22 Feb 2017
TL;DR: Zhang et al. as mentioned in this paper presented an image identification method and device, which consists of obtaining a target image which an image to be identified, performing feature extraction of the target image through a presetting convolution neural network model, and obtaining the attribute vectors of target image, wherein the attribute vector include the probability of a plurality of presetting word attributes in the target images.
43
Patent
Face detection method and device
Song Li,Duan Xu,Zhang Xiangde +2 more
- 03 Apr 2018
TL;DR: Zhang et al. as mentioned in this paper proposed a face detection method and device, which comprises the steps of inputting a to-be-detected image into a first convolution neural network, detecting a face region in the detected face region, adjusting the scale of an (M-1) face candidate window to be the size of an M-th convolutional neural network.
39
Patent
Detection method and apparatus
Yang Ming
- 23 Dec 2015
TL;DR: In this article, a Siamese network is used to detect whether a template image matches a test image, achieving a high accuracy rate of determining whether the test image matches the template image.
26
Patent
Face comparison method based on high-dimensional LBP (Local Binary Patterns) and convolutional neural network feature fusion
Cai Yehe,Huafeng Wang,Huang Jiang,Wenfeng Song,Du Junyi,Lyu Weifeng +5 more
- 04 May 2016
TL;DR: In this article, a face comparison method based on high-dimensional LBP (local binary patterns) and convolutional neural network feature fusion is proposed for real-time face comparison.
25
Patent
Multi-task deep learning network-based training method, system, multi-task deep learning network-based identification method and system
TL;DR: In this paper, a multi-task deep learning network-based training method was proposed, which includes the following steps: the face region of a face image in a training set is obtained; key point detection is performed on the face regions, so that key feature point positions are obtained; affine transformation is performed according to the key feature positions, and the aligned face image is inputted into a trained multitask DNN model.
24
References
Patent
Method for verifying users and updating database, and face verification system using the same
Seok-Cheol Kee
- 22 Apr 2003
TL;DR: In this paper, a method was proposed to reduce degradation of recognition performance due to eye detection errors during face verification and to overcome a problem in that sufficient data to design an optimum feature classifier cannot be obtained during face registration.
125
Patent
Dual-certification face anti-counterfeit method and device
Li Ziqing,Zhiwei Zhang,Lei Zhen,Yi Dong +3 more
- 09 Oct 2013
104
Patent
Feature based classification
Chengjun Liu,Harry Wechsler +1 more
- 31 May 2002
TL;DR: In this article, a novel Gabor Feature Classifier (GFC) is proposed for face recognition, which applies the Enhanced FLD Model (EFM) to an augmented Gabor feature vector derived from the Gabor wavelet transformation of images.
90
Face recognition using the POEM descriptor
TL;DR: A new descriptor, POEM (patterns of oriented edge magnitudes), is described by applying a self-similarity based structure on oriented magnitudes and it is proved that it addresses all three criteria of computational cost, robustness, and discriminative power.
84
The 2013 speaker recognition evaluation in mobile environment
Elie Khoury,B. Vesnicer,Javier Franco-Pedroso,Ricardo Paranhos Velloso Violato,Z. Boulkcnafet,L. M. Mazaira Fernandez,Mireia Diez,J. Kosmala,Houssemeddine Khemiri,T. Cipr,Rahim Saeidi,Manuel Günther,J. Zganec-Gros,R. Zazo Candil,Flávio Olmos Simões,Messaoud Bengherabi,A. Alvarez Marquina,Mikel Peñagarikano,Alberto Abad,M. Boulayemen,Petr Schwarz,D.A. van Leeuwen,Javier Gonzalez-Dominguez,M. Uliani Neto,Elhocine Boutellaa,P. Gómez Vilda,Amparo Varona,Dijana Petrovska-Delacrétaz,Pavel Matejka,Joaquin Gonzalez-Rodriguez,Tiago de Freitas Pereira,Farid Harizi,Luis Javier Rodríguez-Fuentes,L. El Shafey,Marcus de Assis Angeloni,Germán Bordel,Gérard Chollet,Sébastien Marcel +37 more
- 04 Jun 2013
TL;DR: Results in terms of equal error rate, half total error rate (HTER) and detection error trade-off (DET) confirm that the best performing systems are based on total variability modeling, and are the fusion of several sub-systems.
Related Papers (5)
Ma Kun,Li Cheng,Tang Xiaoou +2 more
- 23 Nov 2016
Kyoung Tai Lee
- 05 Nov 2010
Xinman Zhang,Kunlei Jing,Yixuan Dai,Xuebin Xu +3 more
- 01 Dec 2018