Skip to main content
SpringerLink
Log in

A deep hashing method of likelihood function adaptive mapping

  • Original Article
  • Published:
Neural Computing and Applications Aims and scope Submit manuscript

Abstract

In image retrieval, deep-learning-based models combing deep hashing and Bayesian learning have become one of the mainstream approaches. The choice of likelihood functions can significantly affect the performance of existing image retrieval methods that combine deep hashing and Bayesian learning, resulting in issues such as misclassification in single-label datasets and biased label association in multi-label ones. However, it remains to further explore how image retrieval performance can be reliably enhanced through proper likelihood function design. In this paper, we propose a deep adaptive-mapping-based hashing (DAMH) method that enhances image retrieval performance via adjustable likelihood function design. Through strategically re-mapping image samples with low-gradient to high-gradient regions of the likelihood functions, our method both effectively expands the ranges over which inner products used in single-label image retrieval are trained and properly delimits the likelihood functions to prevent multi-label images from being excessively mapped into Hamming sphere(s) of any single class. Furthermore, we design a batch-by-batch optimization method that treats easy and hard samples differently, preventing the gradients of hard samples from being submerged by those of the easy ones during the training process. Our experiments on general-purpose image datasets, including CIFAR10, NUSWIDE and ImageNet100, show that DAMH excels existing peer methods in overall image retrieval performance.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18

Similar content being viewed by others

Data availability

The datasets and model parameter settings generated and/or analyzed during the current study are recorded, documented and made available in our DAMH project repository at Github: https://github.com/q878787/DAMH. In addition to our version of the NUSWIDE and ImageNet100 datasets available at https://github.com/q878787/DAMH/tree/main/data, the original version of the datasets utilized in our current study, which are all publicly available, can be accessed via the following links: 1. CIFAR10 [47] dataset: https://www.cs.toronto.edu/~kriz/cifar.html. 2. NUSWIDE [48] dataset: https://lms.comp.nus.edu.sg/wp-content/uploads/2019/research/nuswide/NUS-WIDE.html. 3. ImageNet100 [49] dataset: https://www.image-net.org/.

Notes

  1. Our source code can be found at https://github.com/q878787/DAMH.

References

  1. Luo X, Chen C, Zhong H, Zhang H, Deng M, Huang J, Hua X (2020) A survey on deep hashing methods. arXiv preprint arXiv:2003.03369

  2. Hecht-Nielsen R (1992) Theory of the backpropagation neural network. Neural networks for perception. Elsevier, Amsterdam, Netherlands, pp 65–93

    Chapter  Google Scholar 

  3. Neal Radford M (2012) Bayesian learning for neural networks, vol 118. Springer, Cham

    MATH  Google Scholar 

  4. Tianfeng C, Draxler Roland R (2014) Root mean square error (RMSE) or mean absolute error (MAE)?-arguments against avoiding RMSE in the literature. Geosci Model Dev 7(3):1247–1250

    Article  Google Scholar 

  5. Kalyan D, Jiming J, Rao JNK (2004) Mean squared error of empirical predictor. Ann Stat 32(2):818–840

    MathSciNet  MATH  Google Scholar 

  6. Hahn G, Lutz SM, Laha N, Lange C (2022) A framework to efficiently smooth L1 penalties for linear regression. bioRxiv, p 2020–09

  7. Golik P, Doetsch P, Ney H (2013) Cross-entropy vs. squared error training: a theoretical and experimental comparison. Interspeech, vol 13. ISCA, Dublin, Ireland, pp 1756–1760

    Google Scholar 

  8. Kline DM, Berardi VL (2005) Revisiting squared-error and cross-entropy functions for training neural network classifiers. Neural Comput Appl 14(4):310–318

    Article  Google Scholar 

  9. Li W-J, Wang S, Kang W-C (2015) Feature learning based deep supervised hashing with pairwise labels. arXiv preprint arXiv:1511.03855

  10. Zhu H, Long M, Wang J, Cao Y (2016) Deep hashing network for efficient similarity retrieval. In: Proceedings of the AAAI conference on artificial intelligence, vol 30

  11. Cao Z, Long M, Wang J, Yu PS (2017) Hashnet: deep learning to hash by continuation. In: Proceedings of the IEEE international conference on computer vision, p 5608–5617

  12. Li Q, Sun Z, He R, Tan T (2017) Deep supervised discrete hashing. arXiv preprint arXiv:1705.10999

  13. Kang R, Cao Y, Long M, Wang J, Yu PS (2019) Maximum-margin hamming hashing. In: Proceedings of the IEEE/CVF international conference on computer vision, p 8251–8260

  14. Cao Z, Sun Z, Long M, Wang J, Yu PS (2018) Deep priority hashing. In: Proceedings of the 26th ACM international conference on multimedia, p 1653–1661

  15. Yang E, Yao D, Cao B, Guan H, Yap P-T, Shen D, Liu M (2020) Deep disentangled hashing with momentum triplets for neuroimage search. International conference on medical image computing and computer-assisted intervention. Springer, Cham, pp 191–201

    Google Scholar 

  16. Cao Y, Long M, Liu B, Wang J (2018) Deep cauchy hashing for hamming space retrieval. In: Proceedings of the IEEE conference on computer vision and pattern recognition, p 1229–1237

  17. Yuan L, Wang T, Zhang X, Tay FEH, Jie Z, Liu W, Feng J (2020) Central similarity quantization for efficient image and video retrieval. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, p 3080–3089

  18. Weiss Y, Torralba A, Fergus R et al (2008) Spectral hashing. In: Proceedings of the 22nd international conference on neural information processing systems (NIPS), vol 1, p 4

  19. Chen S, Cao L, Lin M, Wang Y, Sun X, Wu C, Qiu J, Ji R (2019) Hadamard codebook based deep hashing. arXiv preprint arXiv:1910.09182

  20. Bengio Y, Louradour J, Collobert R, Weston J (2009) Curriculum learning. In: Proceedings of the 26th annual international conference on machine learning, p 41–48

  21. Wang X, Han X, Huang W, Dong D, Scott MR (2019) Multi-similarity loss with general pair weighting for deep metric learning. In: Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, p 5017–5025

  22. Chen W, Liu Y, Wang W, Bakker E, Georgiou T, Fieguth P, Liu L, Lew MS (2021) Deep image retrieval: a survey. arXiv preprint arXiv:2101.11282

  23. Noh H, Araujo A, Sim J, Weyand T, Han B (2017) Large-scale image retrieval with attentive deep local features. In: Proceedings of the IEEE international conference on computer vision, p 3456–3465

  24. Cao B, Araujo A, Sim J (2020) Unifying deep local and global features for image search. European conference on computer vision. Springer, Cham, pp 726–743

    Google Scholar 

  25. Lai H, Pan Y, Liu Y, Yan S (2015) Simultaneous feature learning and hash coding with deep neural networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, p 3270–3278

  26. Yang X, Deng C, Liu T, Tao D (2020) IEEE transactions on pattern analysis and machine intelligence. Heterogeneous graph attention network for unsupervised multiple-target domain adaptation. IEEE, New York

    Google Scholar 

  27. Wang X, Shi Y, Kitani KM (2016) Deep supervised hashing with triplet labels. Asian conference on computer vision. Springer, Cham, pp 70–84

    Google Scholar 

  28. Huang L-K, Chen J, Pan SJ (2019) Accelerate learning of deep hashing with gradient attention. In: Proceedings of the IEEE/CVF international conference on computer vision, p 5271–5280

  29. Liu H, Wang R, Shan S, Chen X (2016) Deep supervised hashing for fast image retrieval. In: Proceedings of the IEEE conference on computer vision and pattern recognition, p 2064–2072

  30. Ji Z, Yao W, Wei W, Song H, Pi H (2019) Deep multi-level semantic hashing for cross-modal retrieval. IEEE Access 7:23667–23674

    Article  Google Scholar 

  31. Xia R, Pan Y, Lai H, Liu C, Yan S (2014) Supervised hashing for image retrieval via image representation learning. In: Twenty-eighth AAAI conference on artificial intelligence, p 2156–2162

  32. Jiang Q-Y, Li W-J (2018) Asymmetric deep supervised hashing. In: Proceedings of the AAAI conference on artificial intelligence, vol 32

  33. Zhang R, Lin L, Zhang R, Zuo W, Zhang L (2015) Bit-scalable deep hashing with regularized similarity learning for image retrieval and person re-identification. IEEE Trans Image Process 24(12):4766–4779

    Article  MathSciNet  MATH  Google Scholar 

  34. Zhao F, Huang Y, Wang L, Tan T (2015) Deep semantic ranking based hashing for multi-label image retrieval. In: Proceedings of the IEEE conference on computer vision and pattern recognition, p 1556–1564

  35. Liu B, Cao Y, Long M, Wang J, Wang J (2018) Deep triplet quantization. In: Proceedings of the 26th ACM international conference on multimedia, p 755–763. Association for computing machinery

  36. Yan X, Zhang L, Li W-J (2017) Semi-supervised deep hashing with a bipartite graph. In: Proceedings of the 26th international joint conference on artificial intelligence, p 3238–3244

  37. Qiu Z, Pan Y, Yao T, Mei T (2017) Deep semantic hashing with generative adversarial networks. In: Proceedings of the 40th international ACM SIGIR conference on research and development in information retrieval, p 225–234

  38. Liu X, Nie X, Yin Y (2019) Mutual linear regression-based discrete hashing. arXiv preprint arXiv:1904.00744

  39. Li Y, Pei W, Gemert J van et al (2019) Push for quantization: deep fisher hashing. arXiv preprint arXiv:1909.00206

  40. Li X, Mengfei X, Jiabo X, Weise T, Zou L, Sun F, Zhize W (2020) Image retrieval using a deep attention-based hash. IEEE Access 8:142229–142242

    Article  Google Scholar 

  41. Zhang P, Zhang W, Li W-J, Guo M (2014) Supervised hashing with latent factor models. In: Proceedings of the 37th international ACM SIGIR conference on research and development in information retrieval, p 173–182

  42. Deng C, Chen Z, Liu X, Gao X, Tao D (2018) Triplet-based deep hashing network for cross-modal retrieval. IEEE Trans Image Process 27(8):3893–3903

    Article  MathSciNet  MATH  Google Scholar 

  43. Yan C, Pang G, Bai X, Shen C, Zhou J, Hancock E (2019) Deep hashing by discriminating hard examples. In: Proceedings of the 27th ACM international conference on multimedia, p 1535–1542

  44. Weisstein EW Disk line picking. https://mathworld.wolfram.com/

  45. Su S, Zhang C, Han K, Tian Y (2018) Greedy hash: towards fast optimization for accurate hash coding in CNN. In: Proceedings of the 32nd international conference on neural information processing systems, p 806–815

  46. Zhu H, Gao S et al (2017) Locality constrained deep supervised hashing for image retrieval. In: Proceedings of the 26th international joint conference on artificial intelligence, p 3567–3573

  47. Krizhevsky A (2009) Learning multiple layers of features from tiny images. ON Canada, Groups at MIT and NYU, Toronto

    Google Scholar 

  48. Chua T-S, Tang J, Hong R, Li H, Luo Z, Zheng Y (2009) Nus-wide: a real-world web image database from national university of singapore. In: Proceedings of the ACM international conference on image and video retrieval, p 1–9

  49. Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M et al (2015) Imagenet large scale visual recognition challenge. Int J Comput Vis 115(3):211–252

    Article  MathSciNet  Google Scholar 

  50. Chatfield K, Simonyan K, Vedaldi A, Zisserman A (2014) Return of the devil in the details: delving deep into convolutional nets. arXiv preprint arXiv:1405.3531

  51. Krizhevsky A, Sutskever I, Hinton GE (2012) ImageNet classification with deep convolutional neural networks. Adv Neural Inf Process Syst 25:1097–1105

    Google Scholar 

  52. Northcutt CG, Athalye A, Mueller J (2021) Pervasive label errors in test sets destabilize machine learning benchmarks. arXiv preprint arXiv:2103.14749

Download references

Acknowledgements

This work is supported by the Guangdong Provincial Foundation for Basic and Applied Basic Research Grant No. \(2021A1515110673\) from the Department of Science and Technology of Guangdong Province, P.R. China. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the funding agency.

Author information

Authors and Affiliations

  1. School of Software, South China Normal University, Nanhai District, Foshan, 528225, Guangdong, People’s Republic of China

    Hai Su, Jianwei Fang, Weixing Liu, Songsen Yu & Huan Yang

Authors
  1. Hai Su
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Jianwei Fang
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Weixing Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Songsen Yu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Huan Yang
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Huan Yang.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Su, H., Fang, J., Liu, W. et al. A deep hashing method of likelihood function adaptive mapping. Neural Comput & Applic 35, 5903–5921 (2023). https://doi.org/10.1007/s00521-022-07962-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00521-022-07962-3

Keywords

Search

Navigation