Realization of a Hybrid Locally Connected Extreme Learning Machine With DeepID for Face Verification

Wong, Shen Yuong; Yap, Keem Siah; Zhai, Qingge; Li, Xiaochao

doi:10.1109/access.2019.2919806

Cited by 13 publications

(9 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For CPS strategy, it was fused with ELDP (called CPS_ELDP in the paper). We test the approaches on four face databases, ORL [40], CALTECH [41], GEORGIA [42], and FACE94 [44] in the experiment, and (R l , 8t) was set (1,8), (2,16), (3,24), (4,32), (5,32), (6,32) and (7,32), respectively. The Nearest Neighbor Classifier (NNC) was chosen for face recognition.…”

Section: A Experimental Set-upmentioning

confidence: 99%

“…Compared with other biometric features, human face contains a lot of detailed information, which is uniqueness for human beings [2], such as eyes, eyebrows and mouths. In addition, because of the low cost of collection and high performance, the face recognition has been commonly applied to surveillance, biometrics, security and other fields [3] in recent years.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Face Recognition Based on Local Gradient Number Pattern and Fuzzy Convex-Concave Partition

Sun

Tang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Face recognition has been deeply studied and widely used in recent years. A novel method, called local gradient number pattern (LGNP), is firstly presented in the paper for face description. For LGNP, the Sobel operator is adopted to extract the local gradient information, and the position of the gray transitions in the local neighborhood is used to form the LGNP code based on the LDP-based methods. Then, the concept of fuzzy convex-concave partition (FCCP) is introduced to fuse the global and regional information based on convex-concave partition (CCP). By the combination of LGNP and FCCP, the proposed descriptor is denoted as FCCP_LGNP. To evaluate the performance of FCCP_LGNP comprehensively, a series of experiments were carried out on four different face databases ORL, CALTECH, GEORGIA, and FACE94, and the results show that FCCP_LGNP is superior to the recent state-of-the-art methods based on hand-crafted features. Even compared with the deep learning methods, VGG16 and ResNet101, the proposed descriptor still shows good performance. INDEX TERMS Face recognition, local gradient number pattern, fuzzy convex-concave partition, LDP-based methods.

show abstract

Section: A Experimental Set-upmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Face Recognition Based on Local Gradient Number Pattern and Fuzzy Convex-Concave Partition

Sun

Tang

et al. 2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…[13]. The recognition accuracy of DeepID [14] proposed by Yi Sun of the Chinese University of Hong Kong on the LFW dataset reached 97.45%. In 2015, Google's FaceNet [15] was better than DeepID, and the accuracy of face recognition on the same face data set reached 99.63%.…”

Section: Introductionmentioning

confidence: 99%

Research on Face Recognition Algorithm Based on Improved Residual Neural Network

Wang¹,

Hou²,

Yiting³

et al. 2021

ACIS

View full text Add to dashboard Cite

The residual neural network is prone to two problems when it is used in the process of face recognition: the first is "overfitting", and the other is the slow or non-convergence problem of the loss function of the network in the later stage of training. In this paper, in order to solve the problem of "overfitting", this paper increases the number of training samples by adding Gaussian noise and salt and pepper noise to the original image to achieve the purpose of enhancing the data, and then we added "dropout" to the network, which can improve the generalization ability of the network. In addition, we have improved the loss function and optimization algorithm of the network. After analyzing the three loss functions of Softmax, center, and triplet, we consider their advantages and disadvantages, and propose a joint loss function. Then, for the optimization algorithm that is widely used through the network at present, that is the Adam algorithm, although its convergence speed is relatively fast, but the convergence results are not necessarily satisfactory. According to the characteristics of the sample iteration of the convolutional neural network during the training process, in this paper, the memory factor and momentum ideas are introduced into the Adam optimization algorithm. This can increase the speed of network convergence and improve the effect of convergence. Finally, this paper conducted simulation experiments on the dataenhanced ORL face database and Yale face database, which proved the feasibility of the method proposed in this paper. Finally, this paper compares the time-consuming and power consumption of network training before and after the improvement on the CMU_PIE database, and comprehensively analyzes their performance.

show abstract

“…More recently, there has been a growing interest for researchers from all over the world have made substantial contributions to ELM variants and applications. Various extensions have been made to the basic ELM to make it more robust and attractive to practitioners, i.e., ELM for representational learning [4], hybridization of locally connected ELM with DeepID for face verification [5], implementation of ELM in Multi Agent System with application to power generation [6]. ELM provides a unified solution for the ''generalized'' SLFNs, which include but not limited to support vector network, traditional neural network and regularized network [7], [8].…”

Section: Introductionmentioning

confidence: 99%

A New Probabilistic Output Constrained Optimization Extreme Learning Machine

Wong

Yap

2020

IEEE Access

Self Cite

View full text Add to dashboard Cite

In near decades machine learning approaches have received overwhelming attention from many researchers for solving problems that cannot be ironed out by traditional approaches. However, most of these approaches produces output that is not equivalent to the probability estimates of how credible and reliable the output can be for each prediction. One widely utilized, highly accorded for generalized performance but non-probabilistic machine learning algorithm is the Extreme Learning Machine (ELM). As with other classification systems, ELM generates outputs that cannot be treated as probabilities. Current literature shows approaches attempt to assimilate probabilistic concept in ELM however their outputs are not equivalent to probabilities. Furthermore, these methods invoke two-stage post processing procedures with iterative learning procedures which are against the salient features of ELM that highlight no iterative operations involved in learning. Hence, we want to probe in this paper the ability of ELM to produce probabilistic output from the original architecture of ELM itself while preserving the merits of ELM without the need for a post processing two-stage procedures to convert the output to probability and eliminates iterative learning to compute output weights. Two methodologies of unified probabilistic ELM framework are presented, i.e., Probabilistic Output Extreme Learning Machine (PO-ELM) and Constrained Optimization Posterior Probabilistic Outputs based Extreme Learning Machine (CPP-POELM). The proposed models are evaluated empirically on several benchmark datasets as well as real world power system applications to demonstrate its validity and efficacy in handling pattern classification problems as well as decision making process. INDEX TERMS Extreme learning machine (ELM), probabilistic outputs, pattern classification, power system applications, confidence threshold.

show abstract

Realization of a Hybrid Locally Connected Extreme Learning Machine With DeepID for Face Verification

Cited by 13 publications

References 43 publications

Face Recognition Based on Local Gradient Number Pattern and Fuzzy Convex-Concave Partition

Face Recognition Based on Local Gradient Number Pattern and Fuzzy Convex-Concave Partition

Research on Face Recognition Algorithm Based on Improved Residual Neural Network

A New Probabilistic Output Constrained Optimization Extreme Learning Machine

Contact Info

Product

Resources

About