度量学习改进语义自编码零样本分类算法

doi:10.13190/j.jbupt.2018-010

北京邮电大学学报 ›› 2018, Vol. 41 ›› Issue (4): 69-75.doi: 10.13190/j.jbupt.2018-010

度量学习改进语义自编码零样本分类算法

陈祥凤^1,2, 陈雯柏^1,2

1. 北京信息科技大学自动化学院, 北京 100192;
2. 北京大学机器感知与智能教育部重点实验室, 北京 100871

收稿日期:2018-01-11 出版日期:2018-08-28 发布日期:2018-10-09
作者简介:陈祥凤(1993-),女,硕士生,E-mail:1057133456@qq.com;陈雯柏(1975-),男,副教授,硕士生导师.
基金资助:
机器感知与智能教育部重点实验室2018年度开放课题（K-2018-08）

Improving Semantic Autoencoder Zero-Shot Classification Algorithm by Metric Learning

CHEN Xiang-feng^1,2, CHEN Wen-bai^1,2

1. College of Automation, Beijing Information Science and Technology University, Beijing 100192, China;
2. MOE Key Laboratory of Machine Perception, Peking University, Beijing 100871, China

Received:2018-01-11 Online:2018-08-28 Published:2018-10-09

摘要/Abstract

摘要： 为改善零样本图像分类中相似度度量方法的鲁棒性，引入了一种用于零样本分类的度量学习方法.该方法由自编码构成，能在特征对齐后的语义嵌入空间中学习到最优的度量函数，用于计算测试样本特征和类标签的语义特征的相似度；然后利用近邻思想预测类别标签，进而避免产生不合适距离函数导致的分类错误.实验结果表明，与传统距离度量的算法相比，所提出的方法降低了识别错误率，在公开数据集AWA、CUB和ImNet-2上的分类准确率分别达到94.7%、63.7%和28.59%；同时表明了语义—视觉的映射方向比相反方向的识别准确率高出2.5%~10.1%.

关键词: 零样本分类, 度量学习, 语义自编码, 语义嵌入空间, 距离函数

Abstract: To improve the robustness of similarity metric method in zero-shot learning, a new metric learning for zero-shot image classification was introduced. It is composed of autoencoders, which can learn the optimal metric function in the feature-aligned semantic embedding space. The similarity between test sample features and the semantic features of the class labels can be calculated by metric function, predicting the label of the class by the neighboring method. Thus, the classification error caused by inappropriate distance function is prevented. Compared with the traditional distance metric algorithm, the experiments show that the proposed method reduces the recognition error rate; the recognition accuracy is improved to 94.7%, 63.7% and 28.59% on the AWA, CUB and ImNet-2 datasets. At the same time, it was confirmed that the recognition accuracy of the semantic-visual mapping direction was 2.5%~10.1% higher than the opposite direction.

Key words: zero-shot classification, metric learning, semantic autoencoder, semantic embedding space, distance function

中图分类号:

TP18

陈祥凤, 陈雯柏. 度量学习改进语义自编码零样本分类算法[J]. 北京邮电大学学报, 2018, 41(4): 69-75.

CHEN Xiang-feng, CHEN Wen-bai. Improving Semantic Autoencoder Zero-Shot Classification Algorithm by Metric Learning[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2018, 41(4): 69-75.

参考文献

[1] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[C]//International Conference on Neural Information Processing Systems.[S. l.]:Curran Associates Inc, 2012:1097-1105.
[2] Lampert C H, Nickisch H, Harmeling S. Attribute-based classification for zero-shot visual object categorization[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(3):453-465.
[3] 乔雪, 彭晨, 段贺, 等. 基于共享特征相对属性的零样本图像分类[J]. 电子与信息学报, 2017, 39(7):1563-1570. Qiao Xue, Peng Chen, Duan He, et al. Zero-shotimage classification based on relative feature of shared features[J]. Journal of Electronics and Information Technology, 2017, 39(7):1563-1570.
[4] 程玉虎, 乔雪, 王雪松. 基于混合属性的零样本图像分类[J]. 电子学报, 2017, 45(6):1462-1468. Cheng Yuhu, Qiao Xue, Wang Xuesong. Zero-shot image classification based on mixed attributes[J]. Chinese Journal of Electronics, 2017, 45(6):1462-1468.
[5] Socher R, Ganjoo M, Manning C D, et al. Zero-shot learning through cross-modal transfer[C]//Advances in Neural Information Processing Systems.[S. l.]:Curran Associates Inc, 2013:935-943.
[6] Fu Y, Hospedales T M, Xiang T, et al. Transductive multi-view zero-shot learning[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(11):2332-2345.
[7] Kodirov E, Xiang T, Gong S. Semantic autoencoder for zero-shot learning[C]//CVPR 2017. Piscataway, NJ:IEEE, 2017:4447-4456.
[8] Li Y, Wang D, Hu H, et al. Zero-shot recognition using dual visual-semantic mapping paths[C]//IEEE Conference on Computer Vision and Pattern Recognition.[S.l.]:IEEE Computer Society, 2017:5207-5215.
[9] Zhang L, Xiang T, Gong S. Learning a deep embedding model for zero-shot learning[EB/OL]. 2016. http://arxiv.org/abs/1611.050B8.
[10] Frome A, Corrado G S, Shlens J, et al. DeViSE:a deep visual-semantic embedding model[C]//International Conference on Neural Information Processing Systems.[S. l.]:Curran Associates Inc, 2013:2121-2129.
[11] 李亚, 王广润, 王青. 基于深度卷积神经网络的跨年龄人脸识别[J]. 北京邮电大学学报, 2017, 40(1):84-88. Li Ya, Wang Guangrun, Wang Qing. Cross-age face recognition based on deep convolutional neural network[J]. Journal of Beijing University of Posts and Telecommunications, 2017, 40(1):84-88.
[12] Liao S, Hu Y, Zhu X, et al. Person reidentification by local maximal occurrence representation and metric learning[EB/OL]. 2014. http://arxiv.org/abs/1406.4216.
[13] Moghaddam B, Jebara T, Pentland A. Bayesian face recognition[J]. Pattern Recognition, 2000, 33(11):1771-1782.
[14] Zhang Z, Saligrama V. Zero-shot learning via joint latent similarity embedding[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. San Francisco:CA, 2016:6034-6042.
[15] Bucher M, Herbin S, Jurie F. Improving semantic embedding consistency by metric learning for zero-shot classiffication[C]//European Conference on Computer Vision. Cham:Springer International Publishing, 2016:730-736.
[16] Shigeto Y, Suzuki I, Hara K, et al. Ridge regression, hubness, and zero-shot learning[C]//The European Conference on Machine Learning and Knowledge Discovery in Databases. Berlin, German:Springer, 2015:135-151.
[17] Ba J L, Swersky K, Fidler S, et al. Predicting deep zero-shot convolutional neural networks using textual descriptions[C]//IEEE International Conference on Computer Vision.[S. l.]:IEEE Computer Society, 2015:4247-4255.
[18] Norouzi M, Mikolov T, Bengio S, et al. Zero-shotlearning by convex combination of semantic embeddings[EB/OL]. 2013. http://arxiv.org/abs/1312.5650.
[19] Fu Z, Xiang T A, Kodirov E, et al. Zero-shot object recognition by semantic manifold distance[C]//Computer Vision and Pattern Recognition.[S. l.]:IEEE, 2015:2635-2644.
[20] Fu Y, Sigal L. Semi-supervised vocabulary-informed learning[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE, 2016:5337-5346.

度量学习改进语义自编码零样本分类算法

Improving Semantic Autoencoder Zero-Shot Classification Algorithm by Metric Learning

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	张天魁蔡昌利骆晓亮朱禹涛. 基于多尺度特征Transformer的细粒度图像分类方法[J]. 北京邮电大学学报, 2023, 46(4): 70-75.
[2]	霍纬纲孟璐郭润夏. 融合重构和对比的多维时间序列表示学习模型[J]. 北京邮电大学学报, 2023, 46(4): 83-90.
[3]	包乾辉文娟石淑珍董萌萍刘雪. 融合句法依存与 BERT-Att-BiLSTM 的属性分类[J]. 北京邮电大学学报, 2023, 46(4): 123-128.
[4]	周雨昊孙哲吴晓非禹可. 基于门控特征融合的中文错别字纠正模型[J]. 北京邮电大学学报, 2023, 46(4): 91-96,122.
[5]	苏兴禹可吴晓非. 基于层次门控交互融合网络的谣言检测方法[J]. 北京邮电大学学报, 2023, 46(4): 97-102.
[6]	张涛马春梅. 面向异构时序数据基于关键帧动态采样的行为识别研究[J]. 北京邮电大学学报, 0, (): 0-0.
[7]	宫岐伟禹可吴晓非. PERC Roberta：基于ERC Roberta的提示学习实现对话情感识别[J]. 北京邮电大学学报, 0, (): 0-0.
[8]	孙艳华史亚会王朱伟李萌司鹏搏. 基于元蒸馏的个性化联邦学习算法[J]. 北京邮电大学学报, 2023, 46(1): 12-18.
[9]	姚日辉陈雯柏陈启丽吴培良. 家庭服务机器人知识图谱的构建与应用[J]. 北京邮电大学学报, 2022, 45(5): 72-78.
[10]	杨杰张书杰. 基于密集YOLOv3的印刷电路板缺陷识别[J]. 北京邮电大学学报, 2022, 45(5): 42-48.
[11]	刘绍华孙靖凯王奕苏刘海波毛天露. 一种多精度交互的行人轨迹预测[J]. 北京邮电大学学报, 2022, 45(5): 1-6.
[12]	刘金桐, 杨国兴, 刘晓鸿, 王光宇. 针对心电图分类模型的平滑攻击算法[J]. 北京邮电大学学报, 2022, 45(4): 44-50.
[13]	黄柳婷, 刘可欣, 牛凯, 常春, 贺志强. 基于深度学习的哮喘患者CT影像黏液栓自动识别[J]. 北京邮电大学学报, 2022, 45(4): 58-63.
[14]	武兴坤, 罗涛, 刘爱军, 杨明, 张文静. 面向主动脉缩窄辅助诊断的神经网络模型设计[J]. 北京邮电大学学报, 2022, 45(4): 1-6.
[15]	谢攀, 彭才静, 何志慧, 张远. 一种融合分组卷积和半监督的睡眠分期方法[J]. 北京邮电大学学报, 2022, 45(4): 7-12.