Random style transfer for person re-identification with one example

Yang Li; Tianshi Wang; Li Liu; Yang Li; Tianshi Wang; Li Liu

doi:10.3934/math.2021277

AIMS Mathematics

2021, Volume 6, Issue 5: 4715-4733. doi: 10.3934/math.2021277

Previous Article Next Article

Research article

Random style transfer for person re-identification with one example

School of Information Science and Engineering, Shandong Normal University, Jinan 250014, China

Received: 09 November 2020 Accepted: 18 February 2021 Published: 24 February 2021
MSC : 68T07, 68U10

Person re-identification with only one labeled image for each identify can not eliminate the style variations of different cameras in the same dataset. In this paper, we propose a random style transfer strategy that randomly transforms the labeled images on the one-example person re-identification task. In this strategy, we focus on twofolds: 1) Randomly transform the camera style of labeled images and unlabeled images during the training stage and 2) use the average feature of labeled data and its camera style transform data to estimate pseudo label on unlabeled data. Notably, our strategy exhibits state-of-the-art performance on large-scale image datasets and its Rank-1 accuracy outperforms the state-of-the-art method by 10.3% points on Market-1501, and 8.4% points on DukeMTMC-reID.
- one-example,
- person re-identification,
- semi-supervised learning,
- style transfer
Citation: Yang Li, Tianshi Wang, Li Liu. Random style transfer for person re-identification with one example[J]. AIMS Mathematics, 2021, 6(5): 4715-4733. doi: 10.3934/math.2021277

Related Papers:

Abstract

Person re-identification with only one labeled image for each identify can not eliminate the style variations of different cameras in the same dataset. In this paper, we propose a random style transfer strategy that randomly transforms the labeled images on the one-example person re-identification task. In this strategy, we focus on twofolds: 1) Randomly transform the camera style of labeled images and unlabeled images during the training stage and 2) use the average feature of labeled data and its camera style transform data to estimate pseudo label on unlabeled data. Notably, our strategy exhibits state-of-the-art performance on large-scale image datasets and its Rank-1 accuracy outperforms the state-of-the-art method by 10.3% points on Market-1501, and 8.4% points on DukeMTMC-reID.

References

[1]	X. Zhu, X. Zhu, M. Li, P. Morerio, S. Gong, Intra-Camera Supervised Person Re-Identification, arXiv, 2002.05046, 2020. Available from: https://arXiv.org/abs/2002.05046.
[2]	Y. Li, Y. Chen, Y. Lin, Y. F. Wang, Cross-Resolution Adversarial Dual Network for Person Re-Identification and Beyond, arXiv, 2002.09274, 2020. Available from: https://arXiv.org/abs/2002.09274.
[3]	S. M. Saquib, A. Schumann, A. Eberle, R. Stiefelhagen, A pose-sensitive embedding for person re-identification with expanded cross neighborhood re-ranking, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2018,420–429, .
[4]	C. Song, Y. Huang, W. Ouyang, L. Wang, Mask-guided contrastive attention model for person re-identification, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2018, 1179–1188.
[5]	Y. Sun, L. Zheng, Y. Yang, Q. Tian, S. Wang, Beyond part models: Person retrieval with refined part pooling (and a strong convolutional baseline), J. Serrin, In: The European Conference on Computer Vision, 2018,480–496.
[6]	Z. Zhong, L. Zheng, Z. Zheng, S. Li, Y. Yang, Camera style adaptation for person re-identification, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2018, 5157–5166.
[7]	S. Liao, Y. Hu, X. Zhu, S. Z. Li, Person re-identification by local maximal occurrence representation and metric learning, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2015, 2197–2206.
[8]	Y. Wu, Y. Lin, X. Dong, Y. Yan, W. Bian, Y. Yang, Progressive learning for person re-identification with one example, IEEE Transactions on Image Processing, 28 (2019), 2872–2881. doi: 10.1109/TIP.2019.2891895
[9]	T. Xiao, H. Li, W. Ouyang, X. Wang, Learning deep feature representations with domain guided dropout for person re-identification, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2016, 1249–1258.
[10]	L. Zhu, Z. Xu, Y. Yang, A. G. Hauptmann, Uncovering the temporal context for video question answering, Int. J. Comput. Vision, 124 (2017), 409–421. doi: 10.1007/s11263-017-1033-7
[11]	C. Deng, Z. Chen, X. Liu, X. Gao, and D. Tao, Triplet-based deep hashing network for cross-modal retrieval, IEEE Trans. Image Proc., 27 (2018), 3893–3903. doi: 10.1109/TIP.2018.2821921
[12]	X. Dong, Y. Yan, M. Tan, Y. Yang, I. W. Tsang, Late fusion via subspace search with consistency preservation, IEEE Trans. Image Proc., 28 (2018), 518–528.
[13]	W. Li, R. Zhao, T. Xiao, X. Wang, Deepreid: Deep filter pairing neural network for person re-identification, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2014,152–159.
[14]	Y. Lee, S. Chen, J. Hwang, Y. Hung, An ensemble of invariant features for person reidentification, IEEE Trans. Circuits Syst. Video Technol., 27 (2016), 470–483.
[15]	Z. Feng, J. Lai, X. Xie, Learning view-specific deep networks for person re-identification, IEEE Trans. Image Proc., 27 (2018), 3472–3483. doi: 10.1109/TIP.2018.2818438
[16]	Y. Lin, X. Dong, L. Zheng, Y. Yan, and Y. Yang, A bottom-up clustering approach to unsupervised person re-identification, J. Serrin, In: The AAAI Conference on Artificial Intelligence, 2019, 8738–8745.
[17]	M. Koestinger, M. Hirzer, P. Wohlhart, P. M. Roth, and H. Bischof, Large scale metric learning from equivalence constraints, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2012, 2288–2295.
[18]	L. Zheng, L. Shen, Scalable person re-identification: A benchmark, J. Serrin, In: The IEEE international conference on computer vision, 2015, 1116–1124.
[19]	Z. Zheng, X. Yang, Z. Yu, L. Zheng, Y. Yang, J. Kautz, Joint discriminative and generative learning for person re-identification, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2019, 2138–2147.
[20]	C. Deng, Z. Chen, X. Liu, X. Gao, D. Tao, Triplet-based deep hashing network for cross-modal retrieval, IEEE Trans. Image Proc., 27 (2018), 3893–3903. doi: 10.1109/TIP.2018.2821921
[21]	Z. Zheng, L. Zheng, Y. Yang, A discriminatively learned cnn embedding for person reidentification, ACM Trans Multimedia Comput., Commun., Appl., 14 (2017), 1–20.
[22]	E. Ahmed, M. Jones, T. K. Marks, An improved deep learning architecture for person re-identification, J. Serrin, In: The IEEE conference on computer vision and pattern recognition,, 2015, 3908–3916.
[23]	T. Huynh-The, CH. Hua, NA. Tu, DS. Kim, Learning 3D spatiotemporal gait feature by convolutional network for person identification, Neurocomputing, 397 (2020), 192–202. doi: 10.1016/j.neucom.2020.02.048
[24]	D. P. Kingma, S. Mohamed, D. J. Rezende, M. Welling, Semi-supervised learning with deep generative models, J. Serrin, In: The Advances in neural information processing systems, 2014, 3581–3589.
[25]	A. Rasmus, M. Berglund, M. Honkala, H. Valpola, T. Raiko, Semi-supervised learning with ladder networks, J. Serrin, In: The Advances in neural information processing systems, 2015, 3546–3554.
[26]	H. Ma, W. Liu, A progressive search paradigm for the internet of things, IEEE MultiMedia, 25 (2017), 76–86.
[27]	X. Dong, L. Zheng, F. Ma, Y. Yang, D. Meng, Few-example object detection with model communication, IEEE Trans. Pattern Anal. Mach. Intell., 41 (2018), 1641–1654.
[28]	L. Qi, L. Wang, J. Huo, Y. Shi, Y. Gao, Progressive Cross-camera Soft-label Learning for Semi-supervised Person Re-identification, IEEE Trans. Circuits Syst. Video Technol., 30 (2020), 2815–2829. doi: 10.1109/TCSVT.2020.2983600
[29]	T. N. Kipf, M. Welling, Semi-supervised classification with graph convolutional networks, arXiv, 1609.02907, 2016. Available from: http://arXiv.org/abs/1609.02907.
[30]	H. Yu, A. Wu, W. Zheng, Cross-view asymmetric metric learning for unsupervised person re-identification, J. Serrin, In: The IEEE international conference on computer vision, 2017,994–1002.
[31]	Z. Liu, D. Wang, H. Lu, Stepwise metric promotion for unsupervised video person re-identification, J. Serrin, In: The IEEE international conference on computer vision, 2017, 2429–2438.
[32]	M. Ye, A. J. Ma, L. Zheng, J. Li, P. C. Yuen, Dynamic label graph matching for unsupervised video re-identification, J. Serrin, In: The IEEE international conference on computer vision, 2017, 5142–5150.
[33]	C. Liu, A. Y. Li, S. Chien, J. Li, Y. Wang, Semantics-Guided Clustering with Deep Progressive Learning for Semi-Supervised Person Re-identification, arXiv, 2010.01148, 2020. Available from: https://arXiv.org/abs/2010.01148.
[34]	P. Peng, X. Tao, Y. Wang, M. Pontil, Y. Tian, Unsupervised cross-dataset transfer learning for person re-identification, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2016, 1306–1315.
[35]	H. Fan, L. Zheng, C. Yan, Y. Yang, Unsupervised person re-identification: Clustering and fine-tuning, ACM Trans. Multimedia Comput., Commun. Appl., 14 (2018), 1–18.
[36]	J. Wang, X. Zhu, S. Gong, W. Li, Transferable joint attribute-identity deep learning for unsupervised person re-identification, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2018, 2275–2284.
[37]	W. Deng, L. Zheng, Image-image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2018,994–1003.
[38]	Z. Zhong, L. Zheng, S. Li, Y. Yang, Generalizing a person retrieval model hetero-and homogeneously, J. Serrin, In: The European Conference on Computer Vision, 2018,172–188.
[39]	S. Xiang, Y. Fu, G. You, T. Liu, Unsupervised domain adaptation through synthesis for person re-identification, J. Serrin, In: The IEEE International Conference on Multimedia and Expo, 2020, 1–6.
[40]	Y. Ge, D. Chen, H. Li, Mutual mean-teaching: Pseudo label refinery for unsupervised domain adaptation on person re-identification, arXiv, 2001.01526, 2020. Available from: https://arXiv.org/abs/2001.01526.
[41]	L. A. Gatys, A. S. Ecker, M. Bethge, Image style transfer using convolutional neural networks, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2016, 2414–2423.
[42]	C. Ledig, L. Theis, Photo-realistic single image super-resolution using a generative adversarial network, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2017, 4681–4690.
[43]	W. Li, R. Zhao, T. Xiao, X. Wang, Deepreid: Deep filter pairing neural network for person re-identification, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2014,152–159.
[44]	L. Ma, Q. Sun, S. Georgoulis, L. Van Gool, B. Schiele, M. Fritz, Disentangled person image generation, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2018, 99–108.
[45]	S. Reed, Z. Akata, X. Yan, L. Logeswaran, B. Schiele, H. Lee, Generative Adversarial Text to Image Synthesis, J. Serrin, In: The International Conference on Machine Learning, 2016, 1060–1069.
[46]	K. P. Dirgantoro, J. M. Lee, D. S. Kim, Generative adversarial networks based on edge computing with blockchain architecture for security system, J. Serrin, In: The International Conference on Artificial Intelligence in Information and Communication, 2020,039–042.
[47]	Z. Zheng, L. Zheng, Y. Yang, Unlabeled samples generated by gan improve the person re-identification baseline in vitro, J. Serrin, In: The IEEE International Conference on Computer Vision, 2017, 3754–3762.
[48]	P. Isola, J. Zhu, T. Zhou, A. Efros, Image-to-image translation with conditional adversarial networks, J. Serrin, In: The IEEE conference on computer vision and pattern recognition, 2017, 1125–1134.
[49]	J. Zhu, T. Park, P. Isola, A. Efros, Unpaired image-to-image translation using cycle-consistent adversarial networks, J. Serrin, In: The IEEE international conference on computer vision, 2017, 2223–2232.
[50]	L. Wei, S. Zhang, W. Gao, Q. Tian, Person transfer gan to bridge domain gap for person re-identification, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2018, 79–88.
[51]	X. Zhang, X. Jing, X. Zhu, F. Ma, Semi-supervised person re-identification by similarity-embedded cycle GANs, Neural Comput. Appl., 32 (2020), 14143–14152. doi: 10.1007/s00521-020-04809-7
[52]	C. Liu, X. Chang, Y. Shen, Unity style transfer for person re-identification, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2020, 6887–6896.
[53]	Y. Chong, C. Peng, J. Zhang, S. Pan, Style transfer for unsupervised domain-adaptive person re-identification, Neural Comput. Appl., 422 (2021), 314–321.
[54]	E. Ristani, F. Solera, R. Zou, R. Cucchiara, C. Tomasi, Performance measures and a data set for multi-target, multi-camera tracking, J. Serrin, In: The European Conference on Computer Vision, 2016, 17–35.
[55]	G. Lisanti, I. Masi, A. D. Bagdanov, B. A. Del, Person re-identification by iterative re-weighted sparse ranking, IEEE transactions on pattern analysis and machine intelligence, 37 (2014), 1629–1642.
[56]	Z. Zhong, L. Zheng, D. Cao, S. Li, Re-ranking person re-identification with k-reciprocal encoding, J. Serrin, In: The IEEE Conference on Computer Vision and Pattern Recognition, 2017, 1318–1327.
[57]	A. Krizhevsky, I. Sutskever, G. E. Hinton, Imagenet classification with deep convolutional neural networks, J. Serrin, In: The Advances in neural information processing systems, 2012, 1097–1105.

Reader Comments

Your name:*

Email:*
© 2021 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)