Abstract
Unsupervised visual tracking has received increasing attention recently. Existing unsupervised visual tracking methods mainly exploit the cycle consistency of sequential images to learn an unsupervised representation for target objects. Due to the small appearance changes between consecutive images, existing unsupervised deep trackers compute the cycle consistency loss over a temporal span to reduce data correlation. However, this causes the learned unsupervised representation not robust to abrupt motion changes as the rich motion dynamics between consecutive frames are not exploited. To address this problem, we propose to contrastively learn cycle consistency over consecutive frames with data augmentation. Specifically, we first use a skipping frame scheme to perform step-by-step cycle tracking for learning unsupervised representation. We then perform unsupervised tracking by computing the contrastive cycle consistency over the augmented consecutive frames, which simulates the challenging scenarios of large appearance changes in visual tracking. This helps us make full use of the valuable temporal motion information for learning robust unsupervised representation. Extensive experiments on large-scale benchmark datasets demonstrate that our proposed tracker significantly advances the state-of-the-art unsupervised visual tracking algorithms by large margins.
J. Zhu—Student.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bachman, P., Hjelm, R.D., Buchwalter, W.: Learning representations by maximizing mutual information across views. In: Advances in Neural Information Processing Systems, pp. 15535–15545 (2019)
Bertinetto, L., Valmadre, J., Henriques, J.F., Vedaldi, A., Torr, P.H.: Fully-convolutional siamese networks for object tracking. In: European Conference on Computer Vision, pp. 850–865 (2016)
Chen, T., Kornblith, S., Norouzi, M., Hinton, G.: A simple framework for contrastive learning of visual representations. arXiv preprint arXiv:2002.05709 (2020)
Cui, Y., Jiang, C., Wang, L., Wu, G.: Target transformed regression for accurate tracking. arXiv preprint arXiv:2104.00403 (2021)
Danelljan, M., Bhat, G., Khan, F.S., Felsberg, M.: Atom: Accurate tracking by overlap maximization. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)
Danelljan, M., Häger, G., Khan, F., Felsberg, M.: Accurate scale estimation for robust visual tracking. In: British Machine Vision Conference (2014)
Danelljan, M., Hager, G., Shahbaz Khan, F., Felsberg, M.: Learning spatially regularized correlation filters for visual tracking. In: The IEEE International Conference on Computer Vision (ICCV) (December 2015)
Fan, H., et al.: Lasot: A high-quality benchmark for large-scale single object tracking. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)
Henriques, J.F., Caseiro, R., Martins, P., Batista, J.: High-speed tracking with kernelized correlation filters. IEEE Trans. Pattern Anal. Mach. Intell. 37(3), 583–596 (2014)
Hu, Q., Zhou, L., Wang, X., Mao, Y., Zhang, J., Ye, Q.: Spstracker: sub-peak suppression of response map for robust object tracking. In: Thirty-Fourth AAAI Conference on Artificial Intelligence (2020)
Huang, Z., Fu, C., Li, Y., Lin, F., Lu, P.: Learning aberrance repressed correlation filters for real-time uav tracking. In: The IEEE International Conference on Computer Vision (ICCV) (October 2019)
Kalal, Z., Mikolajczyk, K., Matas, J.: Forward-backward error: automatic detection of tracking failures. In: 2010 20th International Conference on Pattern Recognition, pp. 2756–2759 (2010)
Kristan, M., Leonardis, A., Matas, J., Felsberg, M., Pflugfelder, R., Cehovin Zajc, L.: The sixth visual object tracking vot2018 challenge results. In: The European Conference on Computer Vision (ECCV) Workshops (September 2018)
Li, B., Wu, W., Wang, Q., Zhang, F., Xing, J., Yan, J.: Siamrpn++: evolution of siamese visual tracking with very deep networks. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)
Li, B., Yan, J., Wu, W., Zhu, Z., Hu, X.: High performance visual tracking with siamese region proposal network. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2018)
Li, X., Liu, S., De Mello, S., Wang, X., Kautz, J., Yang, M.H.: Joint-task self-supervised learning for temporal correspondence. In: Advances in Neural Information Processing Systems, pp. 317–327 (2019)
Russakovsky, O., et al.: Imagenet large scale visual recognition challenge. Int. J. Comput. Vision 115(3), 211–252 (2015)
Tian, Y., Krishnan, D., Isola, P.: Contrastive multiview coding. arXiv preprint arXiv:1906.05849v1 (2019)
Valmadre, J., Bertinetto, L., Henriques, J., Vedaldi, A., Torr, P.H.S.: End-to-end representation learning for correlation filter based tracking. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (July 2017)
Vondrick, C., Shrivastava, A., Fathi, A., Guadarrama, S., Murphy, K.: Tracking emerges by colorizing videos. In: The European Conference on Computer Vision (ECCV) (September 2018)
Wang, N., Song, Y., Ma, C., Zhou, W., Liu, W., Li, H.: Unsupervised deep tracking. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)
Wang, Q., Gao, J., Xing, J., Zhang, M., Hu, W.: Dcfnet: Discriminant correlation filters network for visual tracking. arXiv preprint arXiv:1704.04057 (2017)
Wang, Q., Zhang, L., Bertinetto, L., Hu, W., Torr, P.H.: Fast online object tracking and segmentation: a unifying approach. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)
Wang, X., Jabri, A., Efros, A.A.: Learning correspondence from the cycle-consistency of time. In: The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (June 2019)
Wu, Y., Lim, J., Yang, M.H.: Object tracking benchmark. IEEE Trans. Pattern Anal. Mach. Intell. 37(9), 1834–1848 (2015)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Zhu, J., Ma, C., Jia, S., Xu, S. (2021). Contrastive Cycle Consistency Learning for Unsupervised Visual Tracking. In: Ma, H., et al. Pattern Recognition and Computer Vision. PRCV 2021. Lecture Notes in Computer Science(), vol 13019. Springer, Cham. https://doi.org/10.1007/978-3-030-88004-0_46
Download citation
DOI: https://doi.org/10.1007/978-3-030-88004-0_46
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-88003-3
Online ISBN: 978-3-030-88004-0
eBook Packages: Computer ScienceComputer Science (R0)