Pairwise Costs in Semisupervised Discriminant Analysis for Face Recognition
Pairwise Costs in Semisupervised Discriminant Analysis for Face Recognition
复制标题
人脸识别半监督判别分析中的成对成本
DOI：
10.1109/tifs.2014.2343833
复制
发表时间：
2014-10
影响因子：
6.8
通讯作者：
Chen, Yinjuan
中科院分区：
文献类型：
--
作者：
Wan, Jianwu;Yang, Ming;Gao, Yang;Chen, Yinjuan
近年来，人脸识别被认为是一个代价敏感的学习问题。已经提出了许多成本敏感的分类器。然而，对代价敏感降维的研究还没有得到足够的重视，特别是代价敏感的半监督降维。据我们所知，成本敏感的半监督判别分析（CS3 DA）可能是第一个工作。CS3 DA首先使用稀疏表示来推断未标记样本的软标签，然后通过将误分类成本并入标记和未标记数据来学习投影方向。虽然CS3 DA减少了误分类损失，但它有两个主要缺点：1）稀疏性不是人脸识别的特征，因此稀疏近似可能无法提供所需的鲁棒性或性能; 2）CS3 DA未被证明满足最小误分类损失标准。在本文中，我们嵌入成对成本的半监督判别分析（PCSDA）的人脸识别。PCSDA首先使用简单的l2方法来预测未标记数据的标签，然后通过在标记和未标记数据中嵌入成对成本来学习投影方向。与CS3 DA相比，PCSDA有三个主要优点：1）l2方法比稀疏表示方法更精确、鲁棒; 2）证明了CS3 DA仅在人脸数据集中类平衡且无离群点时才能逼近两两贝叶斯风险; 3）考虑人脸识别中的类不平衡问题和离群点，PCSDA逼近两两贝叶斯风险。因此，通过使用PCSDA得到的投影方向可以更有鉴别力，免疫离群值和类不平衡问题。在AR、PIE、ORL和扩展的Yale B数据集上的实验结果证明了PCSDA的有效性。
In recent years, face recognition is being recognized as a cost-sensitive learning problem. Many cost-sensitive classifiers have been proposed. However, no sufficient attention is paid to the research on cost-sensitive dimensionality reduction, especially on the cost-sensitive semisupervised dimensionality reduction. To the best of our knowledge, cost sensitive semisupervised discriminant analysis (CS3DA) may be the first work. CS3DA first uses the sparse representation to infer a soft label for unlabeled sample and then learns the projection direction by incorporating misclassification costs into both labeled and unlabeled data. Although CS3DA reduces the loss of misclassification, it has two major drawbacks: 1) the sparsity is not a feature of face recognition, and therefore sparse approximations may not deliver the robustness or performance desired and 2) CS3DA is not proven to satisfy the minimal misclassification loss criterion. In this paper, we embed pairwise costs in semisupervised discriminant analysis (PCSDA) for face recognition. PCSDA first uses a simple l2 approach to predict the label of unlabeled data, and then learns the projection direction by embedding pairwise costs in both labeled and unlabeled data. Compared with CS3DA, PCSDA has three major advantages: 1) l2 approach is more accurate and robust than sparse representation for face recognition; 2) we prove that CS3DA approximates the pairwise Bayesian risk only when the classes are balanced and without outliers in face data sets; and 3) PCSDA approximates the pairwise Bayesian risk considering the class imbalance problem and outliers in face recognition. Hence, the projection direction obtained by using PCSDA can be more discriminative, immunes to outliers and class imbalance problem. The experimental results on AR, PIE, ORL, and extended Yale B data sets demonstrate the effectiveness of PCSDA.
登录
查看更多内容
DOI：
10.1109/tc.1978.1674981
发表时间：
2015
期刊：
--
影响因子：
--
作者：
通讯作者：
--
DOI：
10.1609/aaai.v24i1.7661
发表时间：
2010-07
期刊：
--
影响因子：
--
作者：
Yu-Feng Li;James T. Kwok;Zhi-Hua Zhou
通讯作者：
Yu-Feng Li;James T. Kwok;Zhi-Hua Zhou
DOI：
10.1109/tpami.2005.55
发表时间：
2005-03-01
影响因子：
23.6
作者：
He, XF;Yan, SC;Zhang, HJ
通讯作者：
Zhang, HJ
DOI：
10.1109/cvpr.2010.5539952
发表时间：
2010-06
期刊：
2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
影响因子：
--
作者：
Di You;Aleix M. Martinez
通讯作者：
Di You;Aleix M. Martinez
DOI：
10.1109/tkde.2002.1000348
发表时间：
2002-05-01
影响因子：
8.9
作者：
Ting, KM
通讯作者：
Ting, KM