Abstract
Person Re-identification involves detecting a person across non-overlapping camera view videos of the Surveillance Network. This paper uses AB-Divergence based Multivariate Statistical Analysis for Person re-identification. AB-Divergence Multivariate analysis (MVA) estimates weights in linear combinations of feature set, that maximizes the divergence between the feature set variables. Divergence can be defined as measure of statistical dependence between person re-identification feature sets probability distributions. Divergence between the feature sets may be greater than zero and may exists, even when the correlation among feature set is zero. Hence AB-Divergence MVA can be better used to derive the weights for person re-identification feature canonical variate pairs than using Canonical Correlation Analysis. By varying the alpha, beta values the accuracy of person re-identification can be tuned efficiently in training phase. It is observed that the subject wise accuracy can also be fine tuned by varying alpha beta parameters during the training phase.
Keywords
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Niki Martinel, Student Member, IEEE, Gian Luca Foresti, Senior Member, IEEE, and Christian Micheloni, Member, IEEE, “Person Reidentification in a Distributed Camera Network Framework,” IEEE Transactions on Cybernetics 2016
Anirban Chakraborty: Nanyang Technological University, Bappaditya Mandal; Hamed Kiani Galoogah, “Person re-identification using multiple first-person-views on wearable devices” IEEE Winter Conference on Applications of Computer Vision (WACV) 2016
Rahul Rama Varior, Student Member, IEEE, Gang Wang, Member, IEEE, Jiwen Lu, Senior Member, IEEE, and Ting Liu, Student Member, IEEE “Learning Invariant Color Features for Person Re-identification”, IEEE Transactions on Image Processing, VOL. 25, NO. 7 2016
Peng Li; Haiyuan Wu; Qian Chen; Chongke Bi, “Person Re-Identification Using Color Enhancing Feature”, 3rd IAPR Asian Conference on Pattern Recognition 2015
Ancong Wu; Wei-Shi Zheng; Jian-Huang Lai, “Depth-based person re-identification”, 3rd IAPR Asian Conference on Pattern Recognition 2015
Javier Ortiz; Sławomir Bąk; Michał Koperski; François Brémond, “Minimizing hallucination in Histogram of Oriented Gradients”, 12th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) 2015.
Federico Pala, Advisor: Giorgio Fumera, “Re-identification and semantic retrieval of pedestrians in video surveillance scenarios,” Curriculum: ING-INF/05 Computer Engineering Ph.D. in Electronic and Computer Engineering Dept. of Electrical and Electronic Engineering University of Cagliari 2016.
Bingpeng Ma, Yu Su, Frédéric Jurie, “Covariance descriptor based on bio-inspired features for person re-identification and face verification,” Journals-elsevier Image and Vision Computing 2014
Le An, Member, IEEE, Mehran Kafai, Member, IEEE, Songfan Yang, Member, IEEE, and Bir Bhanu, Fellow, IEEE, “Person Reidentification With Reference Descriptor,” IEEE Transactions on Circuits and Systems using Video technology, VOL. 26, NO. 4 2016
Nesrine Abid; Walid Ayedi; Mohamed Abid; Ahmed Chiheb Ammeri “SW/HW Implementation of Image Covariance Descriptor For Person Detection Systems”, 1st International Conference on Advanced Technologies for Signal and Image Processing—ATSIP’ 2014
Giuseppe Lisanti, Svebor Karaman, Iacopo Masi, Alberto Del Bimbo, “Multi Channel-Kernel Canonical Correlation Analysis for Cross-View Person Re-Identification”, IEEE Transactions on Information Forensics and Security, Cornell University Library 2016
Le An; Songfan Yang; Bir Bhanu, “Person Re-Identification by Robust Canonical Correlation Analysis,” IEEE Signal Processing Letters, VOL. 22, NO. 8 2015
Sławomir Bąk; Marco San Biagio; Ratnesh Kumar; Vittorio Murino; François Brémond, “Exploiting Feature Correlations by Brownian Statistics for People Detection and Recognition,” IEEE Transactions on Systems Man and Cybernetics 2016
Chang Tian; Mingyong Zeng; Zemin Wu, “Person Re-Identification Based on Spatiogram Descriptor and Collaborative Representation,” IEEE Signal Processing Letters 2015
Liang Zheng; Liyue Shen; Lu Tian; Shengjin Wang; Jingdong Wang; Qi Tian, “Scalable Person Re-identification: A Benchmark”, IEEE International Conference on Computer Vision 2015
Dapeng Tao, Lianwen Jin, Member, IEEE, Yongfei Wang, and Xuelong Li, Fellow, IEEE (2015), “Person Reidentification by Minimum Classification Error-Based KISS Metric Learning,” IEEE transactions on Cybernetics, VOL. 45, NO. 2 2015
Annan Li, Member, IEEE, Luoqi Liu, Kang Wang, Si Liu, and Shuicheng Yan, Senior Member, IEEE “Clothing Attributes Assisted Person Reidentification,” IEEE Transactions on Circuit and Systems for Video technology, VOL. 25, NO. 5, MAY 2015 869
Giuseppe Lisanti, Iacopo Masi, Andrew D. Bagdanov, Member, IEEE, and Alberto Del Bimbo, Member, IEEE, “Person Re-Identification by Iterative Re-Weighted Sparse Ranking,” IEEE Transactions on Pattern Analysis and Machine Intelligence, VOL. 37, NO. 8 2015
Qin Zhou; Shibao Zheng, Hua Yang;, Yu Wang; and Hang Su, “Joint instance and feature importance re-weighting for person reidentification,” IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) Pages: 1546–1550 2016
Xinxing Xu, Wen Li, Member, IEEE, and Dong Xu, Senior Member, IEEE, “Distance Metric Learning Using Privileged Information for Face Verification and Person Re-Identification”, IEEE Transactions on Neural Network and Learning System, VOL. 26, NO. 12 2015
T. Ojala, M. Pietikainen, T. Maenpaa, “Multiresolution gray-scale and rotation invariant texture classification with local binary patterns”, IEEE Transactions on Pattern Analysis and Machine Intelligence, Jul 2002
A. Mandal, A. Cichocki, “Non-linear Canonical Analysis Using AB-divergence”, Entropy, 2013 Vol. 15, 2788–2804
Zheng, Liang and Bie, Zhi and Sun, Yifan and Wang, Jingdong and Su, Chi and Wang, Shengjin and Tian, Qi, “MARS: A Video Benchmark for Large-Scale Person Re-identification”, Springler, European Conference on Computer 2016
Acknowledgements
We thank Computer Vision and Image Processing Lab, Madras Institute of Technology, Anna University, Chennai, India and Center for Development of Advanced Computing, Chennai, India for facilitating the project implementation.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Harikrishnan, V.S., Sowmiya, D., Anandhakumar, P. (2018). AB Divergence for Fine Tuning Subject Wise Person Re-Identification Performance. In: Chaudhuri, B., Kankanhalli, M., Raman, B. (eds) Proceedings of 2nd International Conference on Computer Vision & Image Processing . Advances in Intelligent Systems and Computing, vol 704. Springer, Singapore. https://doi.org/10.1007/978-981-10-7898-9_12
Download citation
DOI: https://doi.org/10.1007/978-981-10-7898-9_12
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-10-7897-2
Online ISBN: 978-981-10-7898-9
eBook Packages: EngineeringEngineering (R0)