Proceedings

Fast Vehicle Identification in Surveillance via Ranked Semantic Sampling Based Embedding

Download BibTeX Cite

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence. Main track. Pages 3697-3703.

https://doi.org/10.24963/ijcai.2018/514

© 2018 International Joint Conferences on Artificial Intelligence

Abstract

Identifying vehicles across cameras in traffic surveillance is fundamentally important for public safety purposes. However, despite some preliminary work, the rapid vehicle search in large-scale datasets has not been investigated. Moreover, modelling a view-invariant similarity between vehicle images from different views is still highly challenging. To address the problems, in this paper, we propose a Ranked Semantic Sampling (RSS) guided binary embedding method for fast cross-view vehicle Re-IDentification (Re-ID). The search can be conducted by efficiently computing similarities in the projected space. Unlike previous methods using random sampling, we design tree-structured attributes to guide the mini-batch sampling. The ranked pairs of hard samples in the mini-batch can improve the convergence of optimization. By minimizing a novel ranked semantic distance loss defined according to the structure, the learned Hamming distance is view-invariant, which enables cross-view Re-ID. The experimental results demonstrate that RSS outperforms the state-of-the-art approaches and the learned embedding from one dataset can be transferred to achieve the task of vehicle Re-ID on another dataset.

  1. Cross-view M[1] M
  2. M[M[6] M[7] V1-2 90.7 81.4 76.6 80.3 67.3 53.4 3.6 V3-4 95.7 92.9 89.6 91.9 87.8 78.9 9.0 V1-3 86.6 84.5 76.8 82.1 58.2 53.5 4.9 V3-5 97.7 93.0 92.3 92.5 88.4 86.4 6.2 V1-4 98.8 97.5 96.4 96.4 94.6 93.2 9.8 V4-1 95.8 96.0 93.0 93.9 90.3 89.8 11.4 V1-5 88.6 88.5 73.2 78.7 61.5 52.5 6.6 V4-2 89.3 88.8 76.1 84.5 66.4 59.5 4.9 V2-1 86.3 84.2 76.5 82.9 65.6 55.7 4.7 V4-3 92.0 90.0 84.0 86.2 76.5 76.0 9.5 V2-3 86.0 80.3 77.7 80.2 67.3 62.2 3.7 V4-5 94.7 91.8 85.7 85.5 79.7 75.8 7.6 V2-4 88.8 89.7 83.0 82.4 70.9 61.4 1.2 V5-1 88.7 85.5 71.7 79.3 63.2 56.2 4.2 V2-5 96.6 95.8 94.7 94.9 93.9 92.3 2.5 V5-2 95.0 95.6 91.8 92.8 88.8 88.7 4.4 V3-1 86.4 83.4 71.2 81.5 51.4 57.1 4.4 V5-3 93.2 90.8 86.9 88.8 79.7 82.3 6.0 V3-2 91.1 86.9 74.8 83.1 61.7 59.0 2.7 V5-4 95.1 92.7 79.9 88.0 73.6 78.0 5.6
  3. [Amid and Ukkonen, 2015] Ehsan Amid and Antti Ukko- nen. Multiview triplet embedding: Learning attributes in multiple maps. In ICML, 2015.
  4. [Bronstein and Bronstein, 2010] Michael M. Bronstein and Alexander M. Bronstein. Data fusion through cross- modality metric learning using similarity-sensitive hash- ing. In CVPR, 2010.
  5. [Deng et al., 2009] Jia Deng, Wei Dong, Richard Socher, Li- Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A large-scale hierarchical image database. In CVPR, 2009.
  6. [Ding et al., 2014] Guiguang Ding, Yuchen Guo, and Jile Zhou. Collective matrix factorization hashing for multi- modal data. In CVPR, 2014.
  7. [Fellbaum, 2000] Christiane Fellbaum. Wordnet: An elec- tronic lexical database. Linguistic Society of America, 76(3):706-708, 2000.
  8. [Ferrari and Zisserman, 2007] Vittorio Ferrari and Andrew Zisserman. Learning visual attributes. In NIPS, 2007.
  9. [Frome et al., 2013] Andrea Frome, Greg S. Corrado, Jonathon Shlens, Samy Bengio, Jeffrey Dean, MarcAure- lio Ranzato, and Tomas Mikolov. Devise: A deep visual- semantic embedding model. In NIPS, 2013.
  10. [Hadsell et al., 2006] Raia Hadsell, Sumit Chopra, and Yann LeCun. Dimensionality reduction by learning an invariant mapping. In CVPR, 2006.
  11. [Huang et al., 2016] Chen Huang, Chen Change Loy, and Xiaoou Tang. Local similarity-aware deep feature embed- ding. In NIPS, 2016.
  12. [Hwang and Sigal, 2014] Sung Ju Hwang and Leonid Sigal. A unified semantic embedding: Relating taxonomies and attributes. In NIPS, 2014.
  13. [Krizhevsky et al., 2012] Alex Krizhevsky, Ilya Sutskever, and Geoffrey E. Hinton. Imagenet classification with deep convolutional neural networks. In NIPS, 2012.
  14. [Kukliasnky and Shamir, 2015] Doron Kukliasnky and Ohad Shamir. Attribute efficient linear regression with data- dependent sampling. In ICML, 2015.
  15. [Kumar and Udupa, 2011] Shaishav Kumar and Raghaven- dra Udupa. Learning hash functions for cross-view sim- ilarity search. In IJCAI, 2011.
  16. [LeCun et al., 1989] Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, and L. D. Jackel. Backpropagation applied to handwritten zip code recognition. Neural Computation, 1(4):541 -551, 1989.
  17. [Liu et al., 2016a] Hongye Liu, Yonghong Tian, Yaowei Wang, Lu Pang, and Tiejun Huang. Deep relative distance learning: Tell the difference between similar vehicles. In CVPR, 2016.
  18. [Liu et al., 2016b] Xinchen Liu, Wu Liu1, Tao Mei, and Huadong Ma. A deep learning-based approach to pro- gressive vehicle re-identification for urban surveillance. In ECCV, 2016.
  19. [Lowe, 1999] David G. Lowe. Object recognition from local scale-invariant features. In iccv, 1999.
  20. [Rastegari et al., 2013] Mohammad Rastegari, Jonghyun Choi, Shobeir Fakhraei, Hal Daume III, and Larry S. Davis. Predictable dual-view hashing. In ICML, 2013.
  21. [Schroff et al., 2015] Florian Schroff, Dmitry Kalenichenko, and James Philbin. Facenet: A unified embedding for face recognition and clustering. In CVPR, 2015.
  22. [Shen et al., 2017] Yantao Shen, Tong Xiao, Hongsheng Li, Shuai Yi, and Xiaogang Wang. Learning deep neural net- works for vehicle re-id with visual-spatio-temporal path proposals. In ICCV, 2017.
  23. [Sohn, 2016] Kihyuk Sohn. Improved deep metric learning with multi-class n-pair loss objective. In NIPS, 2016.
  24. [Song et al., 2016] Hyun Oh Song, Yu Xiang, Stefanie Jegelka, and Silvio Savarese. Deep metric learning via lifted structured feature embedding. In CVPR, 2016.
  25. [Szegedy et al., 2015] Christian Szegedy, Wei Liu, Yangqing Jia, Pierre Sermanet, Scott Reed, Dragomir Anguelov, Dumitru Erhan, Vincent Vanhoucke, and Andrew Rabi- novich. Going deeper with convolutions. In CVPR, 2015.
  26. [Ustinova and Lempitsky, 2016] Evgeniya Ustinova and Victor Lempitsky. Learning deep embeddings with histogram loss. In NIPS, 2016.
  27. [van de Weijer et al., 2007] Joost van de Weijer, Cordelia Schmid, Jakob Verbeek, and Diane Larlus. Learning color names for real-world applications. In CVPR, 2007.
  28. [van der Maaten and Hinton, 2008] Laurens van der Maaten and Geoffrey Hinton. Visualizing data using t-sne. Journal of Machine Learning Research, 9:2579-2605, 2008.
  29. [Wang et al., 2017] Zhongdao Wang, Luming Tang, Xihui Liu, Zhuliang Yao, Shuai Yi, Jing Shao, Junjie Yan, Shengjin Wang, Hongsheng Li, and Xiaogang Wang. Ori- entation invariant feature embedding and spatial temporal regularization for vehicle re-identification. In ICCV, 2017.
  30. [Yang et al., 2015] Linjie Yang, Ping Luo, Chen Change Loy, and Xiaoou Tang. A large-scale car dataset for fine- grained categorization and verification. In CVPR, 2015.
  31. [Zapletal and Herout, 2016] Dominik Zapletal and Adam Herout. Vehicle re-identification for automatic video traf- fic surveillance. In CVPR, 2016.
  32. [Zheng and Shao, 2016] Feng Zheng and Ling Shao. Learn- ing cross-view binary identities for fast person re- identification. In IJCAI, 2016.
  33. [Zheng et al., 2016] Feng Zheng, Yi Tang, and Ling Shao. Hetero-manifold regularisation for cross-modal hashing. IEEE Transactions on Pattern Analysis and Machine In- telligence, pages 1059-1071, 2016.

References were extracted automatically from the PDF and may contain errors.

Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards