计算机科学

计算机科学 ›› 2023, Vol. 50 ›› Issue (6): 183-193.doi: 10.11896/jsjkx.220400038

• 数据库&大数据&数据科学 • 上一篇    下一篇

基于锚图分类的在线半监督跨模态哈希

秦亮1, 谢良1, 陈盛双1, 徐海蛟2   

  1. 1 武汉理工大学理学院 武汉 430070
    2 广东第二师范学院计算机学院 广州 510303
  • 收稿日期:2022-04-03 修回日期:2022-09-07 出版日期:2023-06-15 发布日期:2023-06-06
  • 通讯作者: 谢良(whutxl@hotmail.com)
  • 作者简介:(839805126@qq.com)
  • 基金资助:
    广东省自然科学基金(2020A151501212);广州市基础研究计划基础与应用基础研究项目(202102080353);广东省普通高校自然科学类特色创新项目(2019KTSCX117)

Online Semi-supervised Cross-modal Hashing Based on Anchor Graph Classification

QIN Liang1, XIE Liang1, CHEN Shengshuang1, XU Haijiao2   

  1. 1 College of Science,Wuhan University of Technology,Wuhan 430070,China
    2 School of Computer Science,Guangdong University of Education,Guangzhou 510303,China
  • Received:2022-04-03 Revised:2022-09-07 Online:2023-06-15 Published:2023-06-06
  • About author:QIN Liang,born in 1996,postgraduate.His main research interests include machine learning and cross-modal retrie-val.XIE Liang,born in 1987,Ph.D,associate prefessor.His main research interests include multimedia retrieval and machine learning.
  • Supported by:
    Natural Science Foundation of Guangdong Province,China(2020A151501212),Basic and Applied Basic Research Project of Guangzhou Basic Research Teaching Program(202102080353) and Characteristic Innovation Project of Natural Science in General Colleges and Universities in Guangdong Province(2019KTSCX117).

PDF (PC)

295

摘要： 近年来,哈希算法由于其存储成本小、检索速度快的特点,在大规模多媒体数据的高效跨模态检索中受到了广泛关注。现有的跨模态哈希算法大多是有监督和无监督方法,其中有监督方法通常能够获得更好的性能,但在实际应用中要求所有数据都被标记并不具有可行性。此外,这些方法大多数是离线方法,面对流数据的输入需要付出高额训练成本且十分低效。针对上述问题,提出了一种新的半监督跨模态哈希方法——在线半监督锚图跨模态哈希(Online Semi-supervised Anchor Graph Cross-modal Hashing,OSAGCH),构建了半监督锚图跨模态哈希模型,在只有部分数据有标签的情况下,利用正则化锚图预测数据标签,并通过子空间关系学习哈希函数,一步生成统一的哈希码,同时针对流数据输入的情况对该模型进行了在线化学习,使其能够处理流数据。在公共多模态数据集上进行了实验,结果表明所提方法的性能优于其他现有方法。

关键词: 跨模态哈希, 半监督学习, 锚图正则化, 在线学习, 子空间学习

Abstract: In recent years,hashing algorithm have been widely concerned in efficient cross-modal retrieval of large-scale multimedia data due to small storage costs and high retrieval speed.Most of the existing cross-modal hashing algorithms are supervised or unsupervised methods,and supervised methods usually achieve better performance.However,in real world applications,it is not feasible to require all data to be labeled.In addition,most of these methods are offline,which need to pay high training costs and are very inefficient when facing input of large stream data.This paper proposes a new semi-supervised cross-modal hashing me-thod－－online semi-supervised anchor graph cross-modal hashing(OSAGCH),which builds a semi-supervised anchor graph cross-modal hashing model.It uses regularized anchor graphs to predict data labels in the case where only part of the data has labels,and uses subspace relationship learning to learn hash functions,generating a unified hash code by one step.Then the model is expanded to online version for streaming data input,allowing it to process streaming data.Experiments on public multi-modal data sets indicate that the performance of proposed method is superior to other existing methods.

Key words: Cross-modal hashing, Semi-supervised learning, Anchor graph regularization, Online learning, Subspace learning

中图分类号: 

  • TP391.3
[1]WANG J,ZHANG T,SEBE N,et al.A survey on learning to hash [J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,40(4):769-790.
[2]GORISSE D,CORD M,PRECIOSO F.Locality-sensitive ha-shing for chi2 distance [J].IEEE Transactions on Pattern Ana-lysis and Machine Intelligence,2011,34(2):402-409.
[3]MATSUSHITA Y,WADA T.Principal component hashing:an accelerated approximate nearest neighbor search[C]//Advances in Image and Video Technology.Tokyo:Springer,2009:374-385.
[4]GONG Y C,LAZEBNIK S,GORDO A,et al.Iterative quantization:a procrustean approach to learning binary codes for large-scale image retrieval [J].IEEE Transactions on Pattern Analysis & Machine Intelligence,2013,35(12):2916-2929.
[5]LIU L,YU M Y,SHAO L.Unsupervised local feature hashing for image similarity search [J].IEEE Trans.Cybern,2015,46(11):2548-2558.
[6]HEO J P,LEE Y,HE J F,et al,Spherical hashing:binary code embedding with hyperspheres [J].IEEE Transactions on Pattern Analysis Machine Intelligence,2015,37(11):2304-2316.
[7]STRECHA C,BRONSTEIN A,BRONSTEIN M,et al.LDA-Hash:improved matching with smaller descriptors [J].IEEE Trans.Pattern Anal.Mach.Intell,2011,34(1):66-78.
[8]LIU W,WANG J,JI R R,et al.Supervised Hashing with Kernels[C]//IEEE Conference on Computer Vision and Pattern Recognition.Providence:CVPR,2012:2074-2081.
[9]NGUYEN V A,DO M N.Deep learning based supervised ha-shing for efficient image retrieval[C]//IEEE International Conference on Multimedia and Expo.Seattle:ICME,2016:1-6.
[10]WANG J,KUMAR S,CHANG S F.Semi-supervised hashingfor large-scale search [J].IEEE Trans.Pattern Anal.Mach.Intell,2012,34(12):2393-2406.
[11]WU C X,ZHU J K,CAI D,et al.Semi-supervised nonlinearhashing using bootstrap sequential projection learning [J].IEEE Trans.Knowl.Data Eng,2013,25(6):1380-1393.
[12]GAO L L,SONG J K,ZOU F H,et al.Scalable multimedia retrieval by deep learning hashing with relative similarity learning[C]//Proceedings of the 23rd ACM International Conference on Multimedia.New York:ACM,2015:903-906.
[13]ZHOU X C,TIAN X,WANG X Z,et al.Bagging-boosting-based semi-supervised multi-hashing with query-adaptive re-ranking [J].Neurocomputing,2018,275(10):916-923.
[14]ZHANG S F,LI J M,ZHANG B.Pairwise teacher-student network for semi-supervised hashing[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops.New York:IEEE,2019:730-737.
[15]ANCULEF R,MENA F,MACALUSO A,et al.Self-Supervised Bernoulli Autoencoders for Semi-Supervised Hashing[C]//Iberoamerican Congress on Pattern Recognition.Porto:Springer,2021:258-268.
[16]WANG G A,HU Q H,CHENG J,et al.Semi-supervised gene-rative adversarial hashing for image retrieval[C]//Proceedings of the 15th European Conference on Computer Vision.Munich:ECCV,2018:469-485.
[17]YU J,WU X J,KITTLER J.Semi-supervised hashing for semi-paired cross-view retrieval[C]//Proceedings of the 24th International Conference on Pattern Recognition(ICPR).Beijing:IEEE,2018:958-963.
[18]SONG T C,CAI J F,ZHANG T Q,et al.Semi-supervised mani-fold-embedded hashing with joint feature representation and classifier learning [J].Pattern Recognition,2017,68(2):99-110.
[19]XU X,SHEN F M,YANG Y,et al.Learning discriminative binary codes for large-scale cross-modal retrieval [J].IEEE Transactions on Image Processing,2017,26(5):2494-2507.
[20]ETO K,KOUTAKI G,SHIRAI K.Hadamard Coded Discrete Cross Modal Hashing[C]//2018 25th IEEE International Conference on Image Processing(ICIP).Switzerland:IEEE,2018:2007-2011.
[21]YU G X,LIU X W,WANG J,et al.Flexible Cross-Modal Ha-shing [J].IEEE Transactions on Neural Networks and Learning Systems,2020,1(1):1-11.
[22]LIN Z J,DING G G,HAN J G,et al.Cross-view retrieval via probability-based semantics-preserving hashing [J].IEEE Tran-sactions on Cybernetics,2016,47(12):4342-4355.
[23]JIANG Q Y,LI W J.Deep cross-modal hashing[C]//Procee-dings of the IEEE Conference on Computer Vision and Pattern Recognition.Hawaii:CVPR,2017:3232-3240.
[24]HUANG L K,YANG Q,ZHENG W S.Online hashing [J].IEEE Transactions on Neural Networks and Learning Systems,2017,29(6):2309-2322.
[25]LIN M B,JI R R,LIU H,et al.Supervised online hashing via hadamard code-book learning[C]//Proceedings of the 26th ACM International Conference on Multimedia.Seoul Republic:ACM,2018:1635-1643.
[26]LIN M B,JI R R,LIU H,et al.Hadamard matrix guided online hashing [J].International Journal of Computer Vision,2020,128(8):2279-2306.
[27]ZHENG C,ZHU L,ZHANG S,et al.Efficient parameter-freeadaptive multi-modal hashing [J].IEEE Signal Processing Letters,2020,27(9):1270-1274.
[28]LIN T Y,MAIRE M,BELONGIE S,et al.Microsoft coco:Common objects in context[C]//European Conference on Computer vision.Zurich:Springer,Cham,2014:740-755.
[29]HUISKES M J,LEW M S.The mir flickr retrieval evaluation[C]//Proceedings of the 1st ACM International Conference on Multimedia Information Retrieval.Canada:ACM,2008:39-43.
[30]CHUA T S,TANG J,HONG R,et al.Nus-wide:a real-world web image database from national university of singapore[C]//Proceedings of the ACM International Conference on Image and Video Retrieval.Greece:CIVR,2009:1-9.
[31]SONG J,YANG Y,HUANG Z,et al.Effective multiple feature hashing for large-scale near-duplicate video retrieval [J].IEEE Transactions on Multimedia,2013,15(8):1997-2008.
[32]LU X,ZHU L,LI J,et al.Efficient supervised discrete multi-view hashing for large-scale multimedia search [J].IEEE Tran-sactions on Multimedia,2019,22(8):2048-2060.
[33]SCHEFFER T,WROBEL S.Active learning of partially hidden markov models[C]//Proceedings of the ECML/PKDD Workshop on Instance Selection.Freiburg:ECML,2001:1-15.
[34]WANG D,WANG Q,GAO X.Robust and flexible discrete ha-shing for cross-modal similarity search [J].IEEE Transactions on Circuits and Systems for Video Technology,2017,28(10):2703-2715.
[35]WANG D,WANG Q,HE L,et al.Joint and individual matrix factorization hashing for large-scale cross-modal retrieval [J].Pattern Recognition,2020,107(4):7-19.
[36]LIU L,YANG Y,HU M,et al.Index and retrieve multimedia data:Cross-modal hashing by learning subspacerelation[C]//International Conference on Database Systems for Advanced Applications.Australia:Springer,2018:606-621.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发