Advertisement

Random linear interpolation data augmentation for person re-identification

  • Zhi Li
  • Jun Guo
  • Wenli Jiao
  • Pengfei XuEmail author
  • Baoying Liu
  • Xiaowei Zhao
Article
  • 43 Downloads

Abstract

Person Re-Identification (person re-ID) is an image retrieval task which identifies the same person in different camera views. Generally, a good person re-ID model requires a large dataset containing over 100000 images to reduce the risk of over-fitting. Most current handcrafted person re-ID datasets, however, are insufficient for training a learning model with high generalization ability. In addition, the lacking of images with various levels of occlusion is still remaining in most existing datasets. Motivated by these two problems, this paper proposes a new data augmentation method called Random Linear Interpolation that can enlarge the sizes of person re-ID datasets and improve the generalization ability of the learning model. The key enabler of our approach is generating fused images by interpolating pairs of original images. In other words, the innovation of the proposed approach is considering data augmentation between two random samples. Plenty of experimental results demonstrates that the proposed method is effective to improve baseline models. On Market1501 and DukeMTMC-reID datasets, our approach can achieve 92.71% and 82.19% rank-1 accuracy, respectively.

Keywords

Person re-identification Data augmentation Linear interpolation 

Notes

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (Program No. 61702415, No.61502387), Natural Science Basic Research Plan in Shaanxi Province of China (Program No.2017JM6056,2016JQ6029) and Talent Support Project of Science Association in Shaanxi Province (Program No. 20180108).

References

  1. 1.
    Bai S, Bai X (2016) Sparse contextual activation for efficient visual re-ranking. IEEE Trans Image Process 25(3):1056–1069MathSciNetCrossRefGoogle Scholar
  2. 2.
    Barbosa IB, Cristani M, Caputo B, Rognhaugen A, Theoharis T (2017) Looking beyond appearances: synthetic training data for deep cnns in re-identification. Computer Vision & Image UnderstandingGoogle Scholar
  3. 3.
    Boroujeni FR, Wang S, Li Z, West N, Stantic B, Yao L, Long G (2018) Trace ratio optimization with feature correlation mining for multiclass discriminant analysis. In: The AAAI conference on artifical intelligenceGoogle Scholar
  4. 4.
    Chang X, Nie F, Wang S, Yang Y, Zhou X, Zhang C (2016) Compound rank- k projections for bilinear analysis. IEEE Transactions on Neural Networks & Learning Systems 27(7):1502–1513MathSciNetCrossRefGoogle Scholar
  5. 5.
    Chapelle O, Weston J (2011) Vicinal risk minimization. Advances in Neural Information Processing Systems, pp 416–422Google Scholar
  6. 6.
    Cheng D, Chang X, Liu L, Hauptmann AG, Gong Y, Zheng N (2017) Discriminative dictionary learning with ranking metric embedded for person re-identification. In: Twenty-sixth international joint conference on artificial intelligence, pp 964–970Google Scholar
  7. 7.
    Datta R, Joshi D, Li J, Wang JZ (2008) Image retrieval: ideas, influences, and trends of the new age. ACM Comput Surv 40(2):1–60CrossRefGoogle Scholar
  8. 8.
    Felzenszwalb P, Mcallester D, Ramanan D (2008) A discriminatively trained, multiscale, deformable part model. Cvpr 8:1–8Google Scholar
  9. 9.
    Geng M, Wang Y, Xiang T, Tian Y (2016) Deep transfer learning for person re-identification. Computer Vision & Image UnderstandingGoogle Scholar
  10. 10.
    Goodfellow IJ, Warde-Farley D, Mirza M, Courville A, Bengio Y (2013) Maxout networks. Computer Science pp 1319–1327Google Scholar
  11. 11.
    He K, Zhang X, Ren S, Sun J (2015) Deep residual learning for image recognition. pp 770–778Google Scholar
  12. 12.
    Hermans A, Beyer L, Leibe B (2017) In defense of the triplet loss for person re-identificationGoogle Scholar
  13. 13.
    Hirzer M, Roth PM, Bischof H (2012) Person re-identification by efficient impostor-based metric learning. In: IEEE Ninth international conference on advanced video and signal-based surveillance, pp 203–208Google Scholar
  14. 14.
    Hirzer M, Roth PM, Stinger M, Bischof H (2012) Relaxed pairwise learned metric for person re-identification. In: European conference on computer vision, pp 780–793Google Scholar
  15. 15.
    Huang G, Liu Z, Maaten LVD, Weinberger KQ (2017) Densely connected convolutional networks. In: IEEE Conference on computer vision and pattern recognition, pp 2261–2269Google Scholar
  16. 16.
    Ioffe S, Szegedy C (2015) Batch normalization: accelerating deep network training by reducing internal covariate shift. pp 448–456Google Scholar
  17. 17.
    Kodirov E, Xiang T, Fu Z, Gong S (2007) Person re-identification by unsupervised ł1 graph learning. Hydrobiologia 415(11):35–40Google Scholar
  18. 18.
    Kodirov E, Xiang T, Gong S (2015) Dictionary learning with iterative laplacian regularisation for unsupervised person re-identification. In: British machine vision conference, pp 44.1–44.12Google Scholar
  19. 19.
    Kreutz-Delgado K, Murray JF, Rao BD, Engan K, Lee TW, Sejnowski TJ (2003) Dictionary learning algorithms for sparse representation. Neural Comput 15(2):349–396CrossRefGoogle Scholar
  20. 20.
    Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: International conference on neural information processing systems, pp 1097–1105Google Scholar
  21. 21.
    Lei Z, Shen J, Liang X, Cheng Z (2016) Unsupervised topic hypergraph hashing for efficient mobile image retrieval. IEEE Transactions on Cybernetics PP (99):1–14Google Scholar
  22. 22.
    Li J, Wei Y, Liang X, Zhao F, Li J, Xu T, Feng J (2017) Deep attribute-preserving metric learning for natural language object retrieval. In: ACM On multimedia conference, pp 181–189Google Scholar
  23. 23.
    Li Z, Nie F, Chang X, Yang Y (2017) Beyond trace ratio: weighted harmonic mean of trace ratios for multiclass discriminant analysis. IEEE Transactions on Knowledge & Engineering PP(99):1–1Google Scholar
  24. 24.
    Liao S, Hu Y, Zhu X, Li SZ (2015) Person re-identification by local maximal occurrence representation and metric learning. In: Computer vision and pattern recognition, pp 2197–2206Google Scholar
  25. 25.
    Lisanti G, Masi I, Bimbo AD (2014) Matching people across camera views using kernel canonical correlation analysis. pp 1–6Google Scholar
  26. 26.
    Maas AL, Hannun AY, Ng AY (2013) Rectifier nonlinearities improve neural network acoustic modelsGoogle Scholar
  27. 27.
    Roth PM, Hirzer M, K?stinger M, Beleznai C, Bischof H (2014) Mahalanobis distance learning for person re-identification. Springer, LondonGoogle Scholar
  28. 28.
    Roth PM, Wohlhart P, Hirzer M, Kostinger M, Bischof H (2012) Large scale metric learning from equivalence constraints. In: IEEE Conference on computer vision and pattern recognition, pp 2288–2295Google Scholar
  29. 29.
    Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. Computer ScienceGoogle Scholar
  30. 30.
    Sun Y, Zheng L, Deng W, Wang S (2017) Svdnet for pedestrian retrieval. In: IEEE International conference on computer vision, pp 3820–3828Google Scholar
  31. 31.
    Tanner MA, Wong WH (1987) The calculation of posterior distributions by data augmentation. Publ Am Stat Assoc 82(398):528–540MathSciNetCrossRefGoogle Scholar
  32. 32.
    Torralba A, Fergus R, Freeman WT (2008) 80 Million Tiny Images: a Large Data Set for Nonparametric Object and Scene Recognition. IEEE Computer SocietyGoogle Scholar
  33. 33.
    Xiong F, Gou M, Camps O, Sznaier M (2014) Person re-identification using kernel-based metric learning methods. In: European conference on computer vision, pp 1–16Google Scholar
  34. 34.
    Xu B, Wang N, Chen T, Li M (2015) Empirical evaluation of rectified activations in convolutional network. Computer ScienceGoogle Scholar
  35. 35.
    Zhang L, Xiang T, Gong S (2016) Learning a discriminative null space for person re-identification. In: Computer vision and pattern recognition, pp 1239–1248Google Scholar
  36. 36.
    Zheng L, Shen L, Tian L, Wang S, Wang J, Tian Q (2015) Scalable person re-identification: a benchmark. In: IEEE international conference on computer vision, pp 1116–1124Google Scholar
  37. 37.
    Zheng L, Yang Y, Hauptmann AG (2016) Person re-identification: past present and futureGoogle Scholar
  38. 38.
    Zheng Z, Zheng L, Yang Y (2017) Pedestrian alignment network for large-scale person re-identificationGoogle Scholar
  39. 39.
    Zheng Z, Zheng L, Yang Y (2017) Unlabeled samples generated by gan improve the person re-identification baseline in vitro. pp 3774–3782Google Scholar
  40. 40.
    Zhong Z, Zheng L, Cao D, Li S (2017) Re-ranking person re-identification with k-reciprocal encoding. pp 3652–3661Google Scholar
  41. 41.
    Zhong Z, Zheng L, Kang G, Li S, Yang Y (2017) Random erasing data augmentationGoogle Scholar
  42. 42.
    Zhu L, Huang Z, Liu X, He X, Sun J, Zhou X (2017) Discrete multimodal hashing with canonical views for robust mobile landmark search. IEEE Trans Multimedia 19(9):2066–2079CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2018

Authors and Affiliations

  • Zhi Li
    • 1
  • Jun Guo
    • 1
  • Wenli Jiao
    • 1
  • Pengfei Xu
    • 1
    Email author
  • Baoying Liu
    • 1
  • Xiaowei Zhao
    • 1
  1. 1.School of Information Science and TechnologyNorthwest UniversityXianChina

Personalised recommendations