Abstract
This paper proposes a novel and efficient re-ranking technque to solve the person re-identification problem in the surveillance application. Previous methods treat person re-identification as a special object retrieval problem, and compute the retrieval result purely based on a unidirectional matching between the probe and all gallery images. However, the correct matching may be not included in the top-k ranking result due to appearance changes caused by variations in illumination, pose, viewpoint and occlusion. To obtain more accurate re-identification results, we propose to reversely query every gallery person image in a new gallery composed of the original probe person image and other gallery person images, and revise the initial query result according to bidirectional ranking lists. The behind philosophy of our method is that images of the same person should not only have similar visual content, refer to content similarity, but also possess similar k-nearest neighbors, refer to context similarity. Furthermore, the proposed bidirectional re-ranking method can be divided into offline and online parts, where the majority of computation load is accomplished by the offline part and the online computation complexity is only proportional to the size of the gallery data set, which is especially suited to the real-time required video investigation task. Extensive experiments conducted on a series of standard data sets have validated the effectiveness and efficiency of our proposed method.
Similar content being viewed by others
Notes
The bidirectional ranks of two different person images may be not symmetric, which means even if person B is the top-1 nearest neighbor of person A, there can have another person C closest to B but far away from A.
References
Ali S, Javed O, Haering N, Kanade T (2010) Interactive retrieval of targets for wide area surveillance. In: Proceedings of the international conference on multimedia, pp 895–898
Baltieri D, Vezzani R, Cucchiara R (2011) 3dpes: 3D people dataset for surveillance and forensics. In: Proceedings of the 1st international ACM workshop on multimedia access to 3D human objects
Dikmen M, Akbas E, Huang TS, Ahuja N (2010) Pedestrian recognition with a learned metric. In: ACCV
Fischler MA, Bolles RC (1981) Random sample consensus: a paradigm for model fitting with applications to image analysis and automated cartography. Commun ACM 24(6):381–395
Friedkin NE (1998) A structural theory of social influence. Cambridge University Press
Frey BJ, Dueck D (2007) Clustering by passing messages between data points. Science 315(5814):972–976
Farenzena M, Bazzani L, Perina A, Murino V, Cristani M (2010) Person re-identification by symmetry-driven accumulation of local features. In: CVPR, pp 2360–2367
Gheissari N, Sebastian TB, Tu PH, Rittscher J, Hartley R (2006) Person re-identification using spatiotemporal appearance. In: CVPR, vol 2, pp 1528–1535
Gray D, Brennan S, Tao H (2007) Evaluating appearance models for recognition, reacquisition, and tracking. In: PETS
Gong SG, Loy CC, Xiang T (2011) Security and surveillance. In: VAH, part 4, pp 455–472
Hirzer M, Beleznai C, Roth PM, Bischof H (2011) Person re-identification by descriptive and discriminative classification. In: SCIA, LNCS, vol 6688, pp 91–102
Huang J, Yang X, Fang X, Lin W, Zhang R (2011) Integrating visual saliency and consistency for re-ranking image search results. IEEE Trans Multimedia 13(4):653–661
Kostinger M, Hirzer M, Wohlhart P, Roth P, Bischof H (2012) Large scale metric learning from equivalence constraints. In: CVPR, pp 2288–2295
Li W, Wu Y, Mukunoki M et al. (2012) Common-near-neighbor analysis for person re-identification. In: IEEE international conference on image processing (ICIP), pp 1621–1624
Leng Q, Hu R, Liang C, Wang Y, Chen J (2013) Bidirectional ranking for person re-identificationIn: Proceedings of IEEE computer society conference on multimedia and exposure (ICME)
Pedronette D, Guimares C, da S Torres R (2011) Exploiting contextual spaces for image re-ranking and rank aggregation. In: Proceedings of the 1st ACM international conference on multimedia retrieval
Rui Y, Huang TS, Ortega M, Mehrotra S (1998) Relevance feedback: a power tool for interactive content-based image retrieval. IEEE Trans Circuits Syst Video Technol 8(5):644–655
Shen X, Lin Z, Brandt J, Avidan S, Wu Y (2012) Object retrieval and localization with spatially-constrained similarity measure and k-NN re-ranking. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 3013–3020
Wang X, Doretto G, Sebastian T, Rittscher J, Tu P (2007) Shape and appearance context modeling. In: ICCV, pp 1–8
Weinberger KQ, Saul LK (2008) Fast solvers and efficient implementations for distance metric learning. In: ICML
Wu Z, Ke Q, Sun J, Shum HY (2011) Scalable face image retrieval with identity-based quantization and multireference reranking. IEEE Trans Pattern Anal Machine Intell 33(10):1991–2001
Xiang ZJ, Chen Q, Liu Y (2012) Person re-identification by fuzzy space color histogram. Multimed Tools Appl. doi:10.1007/s11042-012-1286-7
You H, Chang E, Li B (2001) NNEW: nearest neighbor expansion by weighting in image database retrieval. In: Proceedings of IEEE international conference multimedia and exposure, pp 245–248
Zheng WS, Gong GS, Tao X (2011) Person re-identification by probabilistic relative distance comparison. In: CVPR, pp 649–656
Zhu C, Wen F, Sun J (2011) A rank-order distance based clustering algorithm for face tagging. In: IEEE conference on computer vision and pattern recognition (CVPR), pp 481–488
Acknowledgments
This work was supported by the National Nature Science Foundation of China (61231015, 61172173, 61303114), the Major Science and Technology Innovation Plan of Hubei Province (2013AAA020), the Guangdong-Hongkong Key Domain Breakthrough Project of China (2012A090200007), the China Postdoctoral Science Foundation funded project (2013M530350), the Specialized Research Fund for the Doctoral Program of Higher Education (20130141120024), the Key Technology R&D Program of Wuhan (2013030409020109) and the President Fund of UCAS, and the Open Project Program of the National Laboratory of Pattern Recognition (NLPR).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Leng, Q., Hu, R., Liang, C. et al. Person re-identification with content and context re-ranking. Multimed Tools Appl 74, 6989–7014 (2015). https://doi.org/10.1007/s11042-014-1949-7
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-014-1949-7