计算机科学 ›› 2018, Vol. 45 ›› Issue (9): 299-302.doi: 10.11896/j.issn.1002-137X.2018.09.050
所属专题: 人脸识别
孙金, 陈若煜, 罗恒利
SUN Jin, CHEN Ruo-yu, LUO Heng-li
摘要: 在大数据时代,图片数量非常巨大,但是具有标签的图片非常少。在学习和研究中,常常需要分类标注图片,而大部分图片都是与人脸相关的,因此人脸标注成为了一种进行图片分类标注的有效方法,但人工标注的成本较大。针对有标签图片数量较少以及人工标注成本较大的问题,提出了在主动学习算法的基础上建立计算人脸类标签后验分布的判别模型的方法。该方法基于马尔可夫随机场和高斯过程,考虑到了样本位置、特征的客观联系,在样本之间加入了匹配约束和非匹配约束,匹配约束表示样本之间具有相同的类标签,非匹配约束表示样本之间具有不同的类标签。实验结果表明,根据判别模型得到的类标签后验分布选择样本进行人工标注,大大提高了分类器的精确度。
中图分类号:
[1]HOI S C H,LYU M R.A Semi-Supervised Active Learning Framework for Image Retrieval[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition.IEEE,2005:302-309. [2]WANG L,CHAN K L,ZHANG Z.Bootstrapping SVM active learning by incorporating unlabeled images for image retrieval[C]∥IEEE Computer Society Conference on Computer Vision and Pattern Recognition.IEEE Computer Society,2003:629-634. [3]GAL Y,ISLAM R,GHAHRAMANI Z.Deep Bayesian Active Learning with Image Data[J].arXiv preprint arXiv:1703.02910,2017. [4]LIU S Y,LIU J F,HUANG Q C,et al.Research on Annotation Technology of Face based on improved AP Clustering Algorithm[J].Intelligent Computer and Application,2011,1(3):35-38.(in Chinese) 刘胜宇,刘家锋,黄庆成,等.基于改进AP聚类算法的人脸标注技术研究[J].智能计算机与应用,2011,1(3):35-38. [5]ZHENG S P,LIU H Y,SUN F M,et al.Face Detection and Annotation Based on a Family Digital Photo Album[J].Journal of Liaoning University of Technology(Natural Science Edition),2016,36(3):160-162.(in Chinese) 郑士鹏,刘海云,孙福明,等.基于家庭数字相册的人脸检测与标注[J].辽宁工业大学学报(自然科学版),2016,36(3):160-162. [6]TONG S,CHANG E.Support vector machine active learning for image retrieval[C]∥ACM International Conference on Multimedia.ACM,2001:107-118. [7]RODRIGUES F,PEREIRA F C,RIBEIRO B.Gaussian process classification and active learning with multiple annotators[C]∥International Conference on International Conference on Machine Learning.JMLR.org,2014:II-433. [8]KAPOOR A,GRAUMAN K,URTASUN R,et al.Active Learning with Gaussian Processes for Object Categorization[C]∥IEEE International Conference on Computer Vision.IEEE,2015:1-8. [9]KAPOOR A,HUA G,AKBARZADEH A,et al.Which faces to tag:Adding prior constraints into active learning[C]∥IEEE,International Conference on Computer Vision.IEEE,2009:1058-1065. [10]SUN S,ZHONG P,XIAO H,et al.An MRF Model-Based Active Learning Framework for the Spectral-Spatial Classification of Hyperspectral Imagery[J].IEEE Journal of Selected Topics in Signal Processing,2015,9(6):1074-1088. [11]HU B,MOSER G,SERPICO S B,et al.An active learning heu-ristic using spectral and spatial information for MRF-based classification[C]∥Geoscience and Remote Sensing Symposium.IEEE,2015. [12]GU Y,JIN Z,CHIU S C.Combining Active Learning and Semi-supervised Learning Using Local and Global Consistency[C]∥International Conference on Neural Information Processing.Springer International Publishing,2014:215-222. [13]ZHAO L,SUKTHANKAR G,SUKTHANKAR R.Incremental Relabeling for Active Learning with Noisy Crowdsourced Annotations[C]∥IEEE Third International Conference on Privacy,Security,Risk and Trust.IEEE,2011:728-733. [14]HUANG S J,JIN R,ZHOU Z H.Active learning by querying informative and representative examples[C]∥International Conference on Neural Information Processing Systems.Curran Associates Inc.2010:892-900. [15]LIU K,QIAN X,WANG Z Q.The summarization of active learning algorithm[J].Computer Engineering and Application,2012,48(34):1-4.(in Chinese) 刘康,钱旭,王自强.主动学习算法综述[J].计算机工程与应用,2012,48(34):1-4. |
[1] | 周慧, 施皓晨, 屠要峰, 黄圣君. 基于主动采样的深度鲁棒神经网络学习 Robust Deep Neural Network Learning Based on Active Sampling 计算机科学, 2022, 49(7): 164-169. https://doi.org/10.11896/jsjkx.210600044 |
[2] | 侯夏晔, 陈海燕, 张兵, 袁立罡, 贾亦真. 一种基于支持向量机的主动度量学习算法 Active Metric Learning Based on Support Vector Machines 计算机科学, 2022, 49(6A): 113-118. https://doi.org/10.11896/jsjkx.210500034 |
[3] | 张人之, 朱焱. 基于主动学习的社交网络恶意用户检测方法 Malicious User Detection Method for Social Network Based on Active Learning 计算机科学, 2021, 48(6): 332-337. https://doi.org/10.11896/jsjkx.200700151 |
[4] | 王体爽, 李培峰, 朱巧明. 基于数据增强的中文隐式篇章关系识别方法 Chinese Implicit Discourse Relation Recognition Based on Data Augmentation 计算机科学, 2021, 48(10): 85-90. https://doi.org/10.11896/jsjkx.200800115 |
[5] | 董心悦, 范瑞东, 侯臣平. 基于边际概率分布匹配的主动标记分布学习 Active Label Distribution Learning Based on Marginal Probability Distribution Matching 计算机科学, 2020, 47(9): 190-197. https://doi.org/10.11896/jsjkx.200700077 |
[6] | 李翼宏, 刘方正, 杜镇宇. 一种改进主动学习的恶意代码检测算法 Malware Detection Algorithm for Improving Active Learning 计算机科学, 2019, 46(5): 92-99. https://doi.org/10.11896/j.issn.1002-137X.2019.05.014 |
[7] | 赵海燕, 汪静, 陈庆奎, 曹健. 主动学习在推荐系统中的应用 Application of Active Learning in Recommendation System 计算机科学, 2019, 46(11A): 153-158. |
[8] | 吕巨建, 赵慧民, 陈荣军, 李键红. 基于自适应稀疏邻域重构的无监督主动学习算法 Unsupervised Active Learning Based on Adaptive Sparse Neighbors Reconstruction 计算机科学, 2018, 45(6): 251-258. https://doi.org/10.11896/j.issn.1002-137X.2018.06.045 |
[9] | 李昌利, 张琳, 樊棠怀. 基于自适应主动学习与联合双边滤波的高光谱图像分类 Hyperspectral Image Classification Based on Adaptive Active Learning and Joint Bilateral Filtering 计算机科学, 2018, 45(12): 223-228. https://doi.org/10.11896/j.issn.1002-137X.2018.12.037 |
[10] | 李锋,万小强. 基于关联矩阵的短信自动分类 SMS Automatic Classification Based on Relational Matrix 计算机科学, 2017, 44(Z6): 428-432. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.096 |
[11] | 王长宝,李青雯,于化龙. 面向类别不平衡数据的主动在线加权极限学习机算法 Active,Online and Weighted Extreme Learning Machine Algorithm for Class Imbalance Data 计算机科学, 2017, 44(12): 221-226. https://doi.org/10.11896/j.issn.1002-137X.2017.12.040 |
[12] | 翟俊海,臧立光,张素芳. 在线序列主动学习方法 Online Sequential Active Learning Approach 计算机科学, 2017, 44(1): 37-41. https://doi.org/10.11896/j.issn.1002-137X.2017.01.007 |
[13] | 梁喜涛,顾磊. 基于最近邻的主动学习分词方法 Active Learning in Chinese Word Segmentation Based on Nearest Neighbor 计算机科学, 2015, 42(6): 228-232. https://doi.org/10.11896/j.issn.1002-137X.2015.06.048 |
[14] | 苏赢彬,杜学绘,夏春涛,曹利峰,陈华成. 基于半监督聚类的文档敏感信息推导方法 Sensitive Information Inference Method Based on Semi-supervised Document Clustering 计算机科学, 2015, 42(10): 132-137. |
[15] | 史文丽,郭茂祖,李晋,刘晓燕. SVM与主动学习方法相结合的蛋白质相互作用预测 Protein-protein Interaction Prediction Combining Active Learning with SVM 计算机科学, 2014, 41(2): 82-86. |
|