© The Institution of Engineering and Technology
Unlike the doublet or triplet constraints, a novel kernelised reference-wise metric learning is proposed by constructing reference-wise constraints, which contain similarity information of each sample to all reference samples. After selecting several representative training samples as the reference set, the training data are first mapped into a projected space by the reference sample matrix. Then the problem of metric learning is casted as a multi-label classification problem under the reference-wise constraints, in which an l 2-norm regularised least squares canonical correlation analysis can be used. Besides, the formulation is generalised to kernelised version for further boosting the performance. Experiments on two benchmark person re-identification datasets demonstrate that the approach clearly outperforms the state-of-the-art methods.
References
-
-
1)
-
3. Tao, D.P., Jin, L.W., Wang, Y.F., Yuan, Y., Li, X.L.: ‘Person re-identification by regularized smoothing kiss metric learning’, IEEE Trans. Circuits Syst. Video Technol., 2013, 23, (10), pp. 1675–1685 (doi: 10.1109/TCSVT.2013.2255413).
-
2)
-
6. Liao, S., Hu, Y., Zhu, X., Li, S.Z.: ‘Person reidentification by local maximal occurrence representation and metric learning’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Boston, USA, 2015, pp. 2197–2206.
-
3)
-
13. Cheng, D.S., Cristani, M., Stoppa, M., Bazzani, L., Murinon, V.: ‘Custom pictorial structures for re-identification’. Proc. British Machine Vision Conf., Dundee, UK, 2011, pp. 68.1–68.11.
-
4)
-
5. Xiong, F., Gou, M.R., Camps, O., Sznaier, M.: ‘Person re-identification using kernel-based metric learning methods’. Proc. European Conf. Computer Vision, Zurich, Switzerland, 2014, pp. 1–16.
-
5)
-
2. Kostinger, M., Hirzer, M., Wohlhart, P., Roth, P.M., Bischof, H.: ‘Large scale metric learning from equivalence constraints’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Providence, USA, 2012, pp. 2288–2295.
-
6)
-
14. Baltieri, D., Vezzani, R., Cucchiara, R.: ‘3dpes: 3d people dataset for surveillance and forensics’. Proc. Int. ACM Workshop on Multimedia Access to 3D Human Objects, Arizona, USA, 2011, pp. 59–64.
-
7)
-
10. Sun, L., Ji, S., Ye, J.: ‘Canonical correlation analysis for multilabel classification: a least-squares formulation, extensions, and analysis’, IEEE Trans. Pattern Anal. Mach. Intell., 2011, 33, (1), pp. 194–200 (doi: 10.1109/TPAMI.2010.160).
-
8)
-
9. An, L., Kafai, M., Yang, S., Bhanu, B.: ‘Person reidentification with reference descriptor’, IEEE Trans. Circuits Syst. Video Technol., 2015, 26, (4), pp. 776–787 (doi: 10.1109/TCSVT.2015.2416561).
-
9)
-
7. Shi, Y., Bellet, A., Sha, F.: ‘Sparse compositional metric learning’. Proc. AAAI Conf. Artificial Intelligence, Quebe, Canada, 2014.
-
10)
-
8. Wang, F., Zuo, W., Zhang, L., Meng, D., Zhang, D.: ‘A kernel classification framework for metric learning’, IEEE Trans. Neural Netw. Learn., 2015, 26, (9), pp. 1950–1962 (doi: 10.1109/TNNLS.2014.2361142).
-
11)
-
11. Davis, J.V., Kulis, B., Jain, P., Sra, S., Dhillon, I.S.: ‘Information-theoretic metric learning’. Proc. Int. Conf. Machine Learning, Oregon, USA, 2007, pp. 209–216.
-
12)
-
12. Weinberger, K.Q., Saul, L.K.: ‘Distance metric learning for large margin nearest neighbor classification’, J. Mach. Learn. Res., 2009, 10, (1), pp. 207–244.
-
13)
-
4. Mignon, A., Jurie, F.: ‘PCCA: A new approach for distance learning from sparse pairwise constraints’. Proc. IEEE Conf. Computer Vision and Pattern Recognition, Providence, USA, 2012, pp. 2666–2672.
-
14)
-
1. Yao, B.H., Zhao, Z.C., Liu, K.: ‘Metric learning with trace-norm regularization for person re-identification’. Proc. IEEE Conf. Image Processing, Paris, France, 2014, pp. 2442–2446.
http://iet.metastore.ingenta.com/content/journals/10.1049/el.2016.4079
Related content
content/journals/10.1049/el.2016.4079
pub_keyword,iet_inspecKeyword,pub_concept
6
6