access icon free Metric learning by simultaneously learning linear transformation matrix and weight matrix for person re-identification

© The Institution of Engineering and Technology
Received 29/06/2018, Accepted 19/02/2019, Revised 17/10/2018, Published 25/02/2019

Mahalanobis metric learning is one of the most popular methods for person re-identification. Most existing metric learning methods regularly formulate the person re-identification as an unconstrained optimisation problem and the constraints on the Mahalanobis matrix are seldom imposed. In addition, weights are often used to model the relationships between different variables but they often suffer from boundedness caused by their hand-designed feature. Taking the above two disadvantages into consideration, the authors propose a new metric learning method for person re-identification, which formulates the metric learning problem as a constrained optimisation problem by imposing a constraint on the linear transformation matrix. Furthermore, they treat the weights as unknown variables and introduce a weight learning method instead of designing weight intuitively. Finally, they evaluate the proposed method on two challenging person re-identification databases and show that it performs favourably against the state-of-the-art approaches.

Inspec keywords: matrix algebra; learning (artificial intelligence); optimisation; object detection

Other keywords: constrained optimisation problem; linear transformation matrix; unconstrained optimisation problem; Mahalanobis metric learning; metric learning problem; metric learning method; weight learning method; Mahalanobis matrix; person re-identification databases

Subjects: Computer vision and image processing techniques; Knowledge engineering techniques; Signal processing and detection; Algebra; Optimisation techniques; Algebra; Optimisation techniques

References

    1. 1)
      • 13. Varior, R.R., Haloi, M., Wang, G.: ‘Gated Siamese convolutional neural network architecture for human re-identification’. Proc. European Conf. on Computer Vision, Amsterdam, Netherlands, 2016, pp. 791–808.
    2. 2)
      • 16. Li, W., Zhao, R., Xiao, T., et al: ‘DeepReID: deep filter pairing neural network for person re-identification’. Proc. Conf. on Computer Vision and Pattern Recognition, Columbus, USA, 2014, pp. 152–159.
    3. 3)
      • 3. Gray, D., Tao, H.: ‘Viewpoint invariant pedestrian recognition with an ensemble of localized features’. Proc. European Conf. on Computer Vision, Marseille, France, 2008, pp. 262–275.
    4. 4)
      • 15. Ustinova, E., Ganin, Y., Lempitsky, V.: ‘Multi-region bilinear convolutional neural networks for person re-identification’. Proc. Advanced Video and Signal Based Surveillance, Lecce, Italy, 2017, pp. 1–6.
    5. 5)
      • 45. Zhao, R., Ouyang, W., Wang, X.: ‘Learning mid-level filters for person reidentification’. Proc. Computer Vision and Pattern Recognition, Columbus, USA, 2014, pp. 144–151.
    6. 6)
      • 20. Liu, L., Lu, X., Yuan, Y., et al: ‘Person re-identification by bidirectional projection’. Proc. Int. Conf. on Internet Multimedia Computing and Service, New York, NY, USA, 2014, pp. 1–5.
    7. 7)
      • 2. Saghafi, M.A., Hussain, A.: ‘Review of person re-identification techniques’, IET Comput. Vis., 2014, 8, (6), pp. 455–474.
    8. 8)
      • 26. Shi, H., Yang, Y., Zhu, X., et al: ‘Embedding deep metric for person re-identification: A study against large variations’. Proc. European Conf. on Computer Vision, Amsterdam, Netherlands, 2016, pp. 732–748.
    9. 9)
      • 22. Wang, Q., Gao, J., Yuan, Y.: ‘Embedding structured contour and location prior in siamesed fully convolutional networks for road detection’, IEEE Trans. Intell. Transp. Syst., 2018, 19, pp. 230–241.
    10. 10)
      • 23. Vishwakarma, D.K., Upadhyay, S.: ‘A deep structure of person re-identification using multi-level Gaussian models’, arXiv preprint arXiv:180507720, 2018.
    11. 11)
      • 1. Gong, S., Cristani, M., Yan, S., et al: ‘Person re-identification’ (Springer, London, 2014).
    12. 12)
      • 10. Zheng, W.S., Gong, S., Xiang, T.: ‘Person re-identification by probabilistic relative distance comparison’. Proc. Computer Vision and Pattern Recognition, Colorado Springs, USA, 2011, pp. 649–656.
    13. 13)
      • 39. Guillaumin, M., Verbeek, J., Schmid, C.: ‘Is that you? Metric learning approaches for face identification’. Proc. Int. Conf. on Computer Vision, Kyoto, Japan, 2009, pp. 498–505.
    14. 14)
      • 40. Wu, L., Shen, C., van den Hengel, A.: ‘Deep linear discriminant analysis on fisher networks: a hybrid architecture for person re-identification’, Pattern Recognit., 2017, 65, pp. 238–250.
    15. 15)
      • 4. Farenzena, M., Bazzani, L., Perina, A., et al: ‘Person reidentification by symmetry-driven accumulation of local features’. Proc. Computer Vision and Pattern Recognition, Denver, USA, 2010, pp. 2360–2367.
    16. 16)
      • 35. Guo, X.: ‘Robust subspace segmentation by simultaneously learning data representations and their affinity matrix’. Proc. Int. Conf. on Artificial Intelligence, Beijing, China, 2015, pp. 3547–3553.
    17. 17)
      • 34. Lin, Z., Liu, R., Su, Z.: ‘Linearized alternating direction method with adaptive penalty for low-rank representation’. Advances in Neural Information Processing Systems, Granada Spain, 2011, pp. 612–620.
    18. 18)
      • 28. Wu, Z., Li, Y., Radke, R.J.: ‘Viewpoint invariant human re-identification in camera networks using pose priors and subject-discriminative features’, IEEE Trans. Pattern Anal. Mach. Intell., 2015, 37, (5), pp. 1095–1108.
    19. 19)
      • 38. Xiong, F., Gou, M., Camps, O., et al: ‘Person re-identification using kernel-based metric learning methods’. Proc. European Conf. on Computer Vision, Zurich, 2014, pp. 1–16.
    20. 20)
      • 44. Zhao, R., Ouyang, W., Wang, X.: ‘Person re-identification by salience matching’. Proc. Int. Conf. on Computer Vision, Portland, USA, 2013, pp. 2528–2535.
    21. 21)
      • 18. Karanam, S., Gou, M., Wu, Z., et al: ‘A comprehensive evaluation and benchmark for person re-identification: features, metrics, and datasets’, arXiv preprint, 2016.
    22. 22)
      • 41. Su, C., Zhang, S., Xing, J., et al: ‘Deep attributes driven multicamera person re-identification’. European Conf. on computer vision, Amsterdam, Netherlands, 2016, pp. 475–491.
    23. 23)
      • 19. Ma, B., Su, Y., Jurie, F.: ‘Local descriptors encoded by fisher vectors for person re-identification’. European Conf. on Computer Vision, Florence, Italy, 2012, pp. 413–422.
    24. 24)
      • 30. Tao, D., Jin, L., Wang, Y., et al: ‘Person re-identification by regularized smoothing kiss metric learning’, IEEE Trans. Circuits Syst. Video Technol., 2013, 23, (10), pp. 1675–1685.
    25. 25)
      • 36. Zheng, L., Shen, L., Tian, L., et al: ‘Scalable person re-identification: a benchmark’. Proc. IEEE Int. Conf. on Computer Vision, Santiago, Chile, 2015, pp. 1116–1124.
    26. 26)
      • 6. Weinberger, K.Q., Saul, L.K.: ‘Distance metric learning for large margin nearest neighbor classification’, J. Mach. Learn. Res., 2009, 10, pp. 207–244.
    27. 27)
      • 32. Yang, X., Wang, M., Tao, D.: ‘Person re-identification with metric learning using privileged information’, IEEE Trans. Image Process., 2018, 27, (2), pp. 791–805.
    28. 28)
      • 42. Liu, J., Zha, Z.J., Tian, Q., et al: ‘Multi-scale triplet CNN for person re-identification’. Proc. ACM on Multimedia Conf., Amsterdam, The Netherlands, 2016, pp. 192–196.
    29. 29)
      • 21. Wang, Q., Gao, J., Yuan, Y.: ‘A joint convolutional neural networks and context transfer for street scenes labeling’, IEEE Trans. Intell. Transp. Syst., 2018, 19, pp. 1457–1470.
    30. 30)
      • 24. Guo, Y., Cheung, N.M.: ‘Efficient and deep person re-identification using multilevel similarity’, arXiv preprint arXiv:180311353, 2018.
    31. 31)
      • 5. Liao, S., Hu, Y., Zhu, X., et al: ‘Person re-identification by local maximal occurrence representation and metric learning’. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Boston, USA, 2015, pp. 2197–2206.
    32. 32)
      • 31. Dong, H., Lu, P., Zhong, S., et al: ‘Person re-identification by enhanced local maximal occurrence representation and generalized similarity metric learning’, Neurocomputing, 2018, 307, pp. 25–37.
    33. 33)
      • 46. Cheng, D., Gong, Y., Zhou, S., et al: ‘Person re-identification by multi-channel parts-based CNN with improved triplet loss function’. Proc. Computer Vision and Pattern Recognition, Las Vegas, USA, 2016, pp. 1335–1344.
    34. 34)
      • 8. Mignon, A., Jurie, F.: ‘PCCA: a new approach for distance learning from sparse pairwise constraints’. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Providence, USA, 2012, pp. 2666–2672.
    35. 35)
      • 11. Pedagadi, S., Orwell, J., Velastin, S., et al: ‘Local fisher discriminant analysis for pedestrian re-identification’. Proc. Computer Vision and Pattern Recognition, Portland, USA, 2013, pp. 3318–3325.
    36. 36)
      • 9. Köstinger, M., Hirzer, M., Wohlhart, P., et al: ‘Large scale metric learning from equivalence constraints’. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Providence, USA, 2012, pp. 2288–2295.
    37. 37)
      • 29. Matsukawa, T., Okabe, T., Suzuki, E., et al: ‘Hierarchical Gaussian descriptor for person re-identification’. Proc. Computer Vision and Pattern Recognition, Las Vegas, USA, June 2016, pp. 1363–1372.
    38. 38)
      • 14. Ahmed, E., Jones, M., Marks, T.K.: ‘An improved deep learning architecture for person re-identification’. Proc. Conf. on Computer Vision and Pattern Recognition, Boston, USA, June 2015, pp. 3908–3916.
    39. 39)
      • 25. Chen, D., Yuan, Z., Chen, B., et al: ‘Similarity learning with spatial constraints for person re-identification’. Proc. Computer Vision and Pattern Recognition, Las Vegas, USA, 2016, pp. 1268–1277.
    40. 40)
      • 17. Jose, C., Fleuret, F.: ‘Scalable metric learning via weighted approximate rank component analysis’. European Conf. on Computer Vision, Amsterdam, Netherlands, 2016, pp. 875–890.
    41. 41)
      • 33. Zhang, J., Yuan, Y., Wang, Q.: ‘Largest center-specific margin for dimension reduction’. Proc. of Int. Conf. on Acoustics, Speech and Signal Processing, New Orleans, USA, 2017, pp. 2352–2356.
    42. 42)
      • 27. Zhao, R., Ouyang, W., Wang, X.: ‘Unsupervised salience learning for person reidentification’. Proc. Computer Vision and Pattern Recognition, Portland, USA, 2013, pp. 3586–3593.
    43. 43)
      • 37. Li, W., Wang, X.: ‘Locally aligned feature transforms across views’. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, Portland, USA, 2013, pp. 3594–3601.
    44. 44)
      • 7. Davis, J.V., Kulis, B., Jain, P., et al: ‘Information-theoretic metric learning’. Int. Conf. on Machine Learning, Corvallis, Oregon, USA, 2007, pp. 209–216.
    45. 45)
      • 12. Li, Z., Chang, S., Liang, F., et al: ‘Learning locally-adaptive decision functions for person verification’. Proc. Computer Vision and Pattern Recognition, Portland, USA, 2013, pp. 3610–3617.
    46. 46)
      • 43. Li, W., Zhao, R., Wang, X.: ‘Human reidentification with transferred metric learning’. Proc. Asian Conf. on Computer Vision, Daejeon, Korea, 2012, pp. 31–44.
    47. 47)
      • 47. Paisitkriangkrai, S., Shen, C., Van Den Hengel, A.: ‘Learning to rank in person re-identification with metric ensembles’. Proc. Computer Vision and Pattern Recognition, Boston, USA, 2015, pp. 1846–1855.
http://iet.metastore.ingenta.com/content/journals/10.1049/iet-cvi.2018.5402
Loading

Related content

content/journals/10.1049/iet-cvi.2018.5402
pub_keyword,iet_inspecKeyword,pub_concept
6
6
Loading