Siamese multi-layer perceptrons for dimensionality reduction and face identification

Abstract

This paper presents a framework using siamese Multi-layer Perceptrons (MLP) for supervised dimensionality reduction and face identification. Compared with the classical MLP that trains on fully labeled data, the siamese MLP learns on side information only, i.e., how similar of data examples are to each other. In this study, we compare it with the classical MLP on the problem of face identification. Experimental results on the Extended Yale B database demonstrate that the siamese MLP training with side information achieves comparable classification performance with the classical MLP training on fully labeled data. Besides, while the classical MLP fixes the dimension of the output space, the siamese MLP allows flexible output dimension, hence we also apply the siamese MLP for visualization of the dimensionality reduction to the 2-d and 3-d spaces.

A Novel Kinship Verification Method Based on Deep Transfer Learning and Feature Nonlinear Mapping

“…However, it may cause a problem that the projected data may mix together and are not distinguishing for kinship verification. To solve the above problem, siamese multilayer perceptrons [9] is adopted to map the feature vector by nonlinear way.…”

“…The modified triangular similarity metric learning objective function [9] , the constant K is the average value of all the features, the y1i and y2i respectively represent the feature after nonlinear mapping, and the si represents the label (1: similarity, -1: dissimilarity). The gradient function of loss function J is:…”

Guest Editorial: Immersive Audio/Visual Systems

“…In this special issue, Zheng et al [16] focus on recognizing human faces with low dimensional feature representation. They implant the traditional neural network, multi-layer perceptrons (MLP), in a siamese architecture to realize flexible dimensionality reduction but maintain good recognition performance.…”

Discriminative Metric Learning with Deep Forest

“…The numerical experiments illustrate the proposed distance metric algorithm. of supervised distance metric learning is cast into pairwise constraints: the equivalence constraints where pairs of data points that belong to the same classes, and inequivalence constraints where pairs of data points belong to different classes.Metric learning approaches were reviewed in [1,5,14,29]. The basic idea underlying the metric learning solution is that the distance between similar objects should be smaller than the distance between different objects.…”

“…This is only a part of successful applications of Siamese neural networks. Many modifications of Siamese networks have been developed, including fully-convolutional Siamese networks [3], Siamese networks combined with a gradient boosting classifier [15], Siamese networks with the triangular similarity metric [29].A new powerful method, which can be viewed as an alternative to deep neural networks, is the deep forest proposed by Zhou and Feng [30] and called the gcForest. It can be compared with a multilayer neural network structure, but each layer in the gcForest contains many random forests instead of neurons.…”