Back
DOI: 10.17586/1023-5086-2020-87-08-52-57
УДК: 004.272 004.032.26
Method of instance normalization in deep-learning-based models for re-identification
For Russian citation (Opticheskii Zhurnal):
Ященко А.В., Родионов С.А., Потапов А.С. Применение метода экземплярной нормализации в моделях на основе глубокого обучения для задачи повторной идентификации // Оптический журнал. 2020. Т. 87. № 8. С. 52–57. http://doi.org/10.17586/1023-5086-2020-87-08-52-57
Yashchenko A.V., Rodionov S.A., Potapov A.S. Method of instance normalization in deep-learning-based models for re-identification [in Russian] // Opticheskii Zhurnal. 2020. V. 87. № 8. P. 52–57. http://doi.org/10.17586/1023-5086-2020-87-08-52-57
For citation (Journal of Optical Technology):
A. V. Yashchenko, S. A. Rodionov, and A. S. Potapov, "Method of instance normalization in deep-learning-based models for re-identification," Journal of Optical Technology. 87(8), 487-490 (2020). https://doi.org/10.1364/JOT.87.000487
Abstract:
In this work, models for re-identification of pedestrians, based on deep learning, are studied. The models considered here use only images of pedestrians and do not require additional spatiotemporal information that can vary in different instances. A modification of the investigated models is proposed, based on the normalization of feature maps. In the best case, an assessment of the accuracy of the model demonstrates a 10% improvement over unmodified models.
Keywords:
deep learning, re-identification of pedestrians
OCIS codes:
150.1135, 100.4996
References:
1. E. Poongothai and A. Suruliandi, “Colour, texture and shape feature analysis for person re-identification technique,” Adv. Vision Comput. 3(3), 17–26 (2016).
2. M. Jaderberg, K. Simonyan, and A. Zisserman, “Spatial transformer networks,” Adv. Neural Inf. Process. Syst. 28, 2017–2025 (2015).
3. A. Potapov, O. Shcherbakov, and I. Zhdanov, “HyperNets and their application to learning spatial transformations,” in 27th International Conference on Artificial Neural Networks, Rhodes, Greece, October 4–7, 2018, pp. 476–486.
4. S. Sabour, N. Frosst, and G. E. Hinton, “Dynamic routing between capsules,” Adv. Neural Inf. Process. Syst. 30, 3856–3866 (2017).
5. A. Hermans and L. Beyer, “In defense of the triplet loss for person reidentification,” arXiv:1703.07737 (2017).
6. F. Zhang, “View confusion feature learning for person reidentification,” in IEEE International Conference on Computer Vision (ICCV), Seoul, South Korea, October 27–November 2, 2019, pp. 6639–6648.
7. H. Luo and Y. Gu, “Bag of tricks and a strong baseline for deep person re-identification,” arXiv:1903.07071 (2019).
8. A. Potapov, S. Rodionov, and H. Latapie, “Metric embedding autoencoders for unsupervised cross-dataset transfer learning,” arXiv:1807.10591 (2018).
9. S. Rodionov, A. Potapov, and H. Latapie, “Improving deep models of person re-identification for cross-dataset usage,” in Artificial Intelligence Applications and Innovations (2018), pp. 75–84.
10. Z. Zhong and L. Zheng, “Invariance matters: exemplar memory for domain adaptive person re-identification,” in Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach,
California, USA, June 16–20, 2019, pp. 598–607.
11. L. Wei and S. Zhang, “Person transfer GAN to bridge domain gap for person re-identification,” in Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, Utah, USA, June 18–22, 2018, pp. 79–88.
12. D. Ulyanov and A. Vedaldi, “Instance normalization: the missing ingredient for fast stylization,” arXiv:1607.08022 (2016).
13. Z. Zheng and X. Yang, “Joint discriminative and generative learning for person re-identification,” in Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, June 16–20, 2019, pp. 2138–2147.
14. Z. Zhong and L. Zheng, “Camera style adaptation for person reidentification,” in Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, Utah, USA, June 18–22, 2018, pp. 5157–5166.
15. J. Jia and Q. Ruan, “Frustratingly easy person re-identification: generalizing person re-ID in practice,” in 30th British Machine Vision Conference, Cardiff, UK, September 9–12, 2019, pp. 117–131.
16. K. He and X. Zhang, “Deep residual learning for image recognition,” in Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, Nevada, USA, June 26–July 1, 2016, pp. 770–778.
17. S. Ioffe and C. Szegedy, “Batch normalization: accelerating deep network training by reducing internal covariate shift,” in International Conference on Machine Learning Lille, France, July 7–9, 2015, pp. 448–456.
18. J. Deng and W. Dong, “ImageNet: a large-scale hierarchical image database,” in IEEE Conference on Computer Vision and Pattern Recognition, Miami Beach, FL, USA, June 20–21, 2009, pp. 248–255.
19. W. Li and R. Zhao, “DeepReID: deep filter pairing neural network for person re-identification,” in IEEE Conference on Computer Vision and Pattern Recognition Columbus, Ohio, USA, June 24–27, 2014, pp. 152–159.
20. S. Bak and P. Carr, “Domain adaptation through synthesis for unsupervised person re-identification,” in European Conference on Computer Vision (ECCV) (2018), pp. 193–209.
21. L. Zheng and L. Shen, “Scalable person re-identification: a benchmark,” in IEEE International Conference on Computer Vision (ICCV), Santiago, Chile, 2015, pp. 1116–1124.
22. E. Ristani, F. Solera, R. Zou, R. Cucchiara, and C. Tomasi, “Performance measures and a data set for multi-target, multi-camera tracking,” (2016), https://users.cs.duke.edu/~tomasi/papers/ristani/ ristaniBmtt16.pdf.
23. A. Wu, W.-S. Zheng, S. Gong, and J. Lai, “RGB-infrared crossmodality person re-identification,” in IEEE International Conference on Computer Vision, Venice, Italy, October 22–29, 2017, pp. 5390–5399.
24. D. Gray, S. Brennan, and H. Tao, “Evaluating appearance models for recognition, reacquisition, and tracking,” in IEEE International Workshop on Performance Evaluation for Tracking and Surveillance, Rio de Janeiro, Brazil, October 14, 2007.
ru