Research on the Correlation Filter Tracking Model Based on the Deep-Pruned Feature Network
Source abstract
Abstract
Visual tracking is one of the key research fields in computer vision. Based on the combination of correlation filter tracking (CFT) model and deep convolutional neural networks (DCNNs), deep correlation filter tracking (DCFT) has recently become a critical issue in visual tracking because of CFT’s rapidity and DCNN’s better feature representation. However, DCNNs are often complex in structure, which most possibly results in the conflict between the rapidity and accuracy of DCFT. To reduce such conflict, this paper proposes a model mainly including: (1) Based on the pre-pruning network obtained by feature channel importance, an optimal global tracking pruning rate (GTPR) is determined in terms of the contribution of filter channels to tracking response. (2) Based on (GTPR), an alternative convolutional kernel is defined to replace non-important channel kernels, which leads to the further pruning of the feature network. (3) An online updating pruned feature network with a structural similarity index is employed to adapt the model to tracking scene changes. (4) The proposed model was performed on OTB2013; experimental results demonstrate the model can effectively enhance speed with a 45% increment while guaranteeing tracking accuracy, and improve tracking accuracy with a 4% increment when tracking scene changes take place.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Video Surveillance and Tracking Methods · Advanced Vision and Imaging · Advanced Measurement and Detection Methods
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Information Engineering, Minzu University of China, Beijing 100081, China
Archives, Yunnan University, Kunming 650091, China
Funding
National Natural Science Foundation of China · National Key R&D Program of China
References (36)
Show source-provided references
1. Li, C., and Chen, H. (2021, January 12–14). Research on lightweight deep correlation filter tracking algorithm based on fuzzy decision. Proceedings of the 2021 2nd International Conference on Computer Science and Management Technology (ICCSMT), Shanghai, China.10.1109/ICCSMT54525.2021.00076 ↗
2. Chen, H., and Li, C. (2021, January 12–14). Pruning deep feature networks using channel importance propagation. Proceedings of the 2021 2nd International Conference on Computer Science and Management Technology (ICCSMT), Shanghai, China.10.1109/ICCSMT54525.2021.00080 ↗
3. Research on deep correlation filter tracking based on channel importance10.1186/s13634-022-00860-9 ↗
4. Li, C., and Yang, G. Deep correlation filter visual tracking algorithm based on channel importance and target similarity, (unpublished).
5. Mean shift: A robust approach toward feature space analysis10.1109/34.1000236 ↗
6. Shi, J. (1994, January 21–23). Good features to track. Proceedings of the 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.
7. An adaptive color-based particle filter10.1016/S0262-8856(02)00129-4 ↗
8. Bolme, D.S., Beveridge, J.R., Draper, B.A., and Lui, Y.M. (2010, January 13–18). Visual object tracking using adaptive correlation filters. Proceedings of the 23rd IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2010, San Francisco, CA, USA.10.1109/CVPR.2010.5539960 ↗
9. Henriques, J.F., Caseiro, R., Martins, P., and Batista, J. (2012, January 7–13). Exploiting the Circulant Structure of Tracking-by-Detection with Kernels. Proceedings of the 12th European Conference on Computer Vision, Florence, Italy.10.1007/978-3-642-33765-9_50 ↗
10. High-Speed Tracking with Kernelized Correlation Filters10.1109/TPAMI.2014.2345390 ↗
11. Agapito, L., Bronstein, M.M., and Rother, C. (2014, January 6–12). A Scale Adaptive Kernel Correlation Filter Tracker with Feature Integration. Proceedings of the Computer Vision–ECCV 2014 Workshops, Zurich, Switzerland.10.1007/978-3-319-16199-0 ↗
12. Imagenet classification with deep convolutional neural networks10.1145/3065386 ↗
13. Deep residual shrinkage networks for fault diagnosis10.1109/TII.2019.2943898 ↗
14. Danelljan, M., Häger, G., Khan, F., and Felsberg, M. (2014). Accurate Scale Estimation for Robust Visual Tracking, Bmva Press.10.5244/C.28.65 ↗
15. Xin, J., Du, X., and Zhang, J. (2017, January 10–14). Deep learning for robust outdoor vehicle visual tracking. Proceedings of the 2017 IEEE International Conference on Multimedia and Expo (ICME), Hong Kong, China.10.1109/ICME.2017.8019329 ↗
16. Recent advances in efficient computation of deep convolutional neural networks10.1631/FITEE.1700789 ↗
17. Deep visual tracking: Review and experimental comparison10.1016/j.patcog.2017.11.007 ↗
18. Voigtlaender, P., and Leibe, B. (2017). Online adaptation of convolutional neural networks for video object segmentation. arXiv.10.5244/C.31.116 ↗
19. Nam, H., and Han, B. (2016, January 27–30). Learning multi-domain convolutional neural networks for visual tracking. Proceedings of the Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA.10.1109/CVPR.2016.465 ↗
20. Song, Y., Ma, C., Gong, L., Zhang, J., Lau, R.W., and Yang, M. (2017, January 2–29). Crest: Convolutional residual learning for visual tracking. Proceedings of the IEEE International Conference on Computer Vision (ICCV), Venice, Italy.10.1109/ICCV.2017.279 ↗
21. Li, Y., Song, L., Chen, Y., Li, Z., Zhang, X., Wang, X., and Sun, J. (2020, January 13–19). Learning dynamic routing for semantic segmentation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA.10.1109/CVPR42600.2020.00858 ↗
22. Jia, X., De Brabandere, B., Tuytelaars, T., and Gool, L.V. (2016, January 5–10). Dynamic filter networks. Proceedings of the Advances in Neural Information Processing Systems 29: Annual Conference on Neural Information Processing Systems 2016, Barcelona, Spain.
23. Li, H., Kadav, A., Durdanovic, I., Samet, H., and Graf, H.P. (2016). Pruning filters for efficient convnets. arXiv.
24. He, Y., Kang, G., Dong, X., Fu, Y., and Yang, Y. (2018). Soft filter pruning for accelerating deep convolutional neural networks. arXiv.10.24963/ijcai.2018/309 ↗
25. Yang, X., Lu, H., Shuai, H., and Yuan, X.T. (2019, January 8–11). Pruning Convolutional Neural Networks via Stochastic Gradient Hard Thresholding. Proceedings of the Second Chinese Conference, PRCV 2019, Xi’an, China.10.1007/978-3-030-31654-9_32 ↗
26. Liu, Z., Li, J., Shen, Z., Huang, G., Yan, S., and Zhang, C. (2017, January 2–29). Learning efficient convolutional networks through network slimming. Proceedings of the IEEE International Conference on Computer Vision, Venice, Italy.10.1109/ICCV.2017.298 ↗
27. Ye, Y., You, G., Fwu, J., Zhu, X., Yang, Q., and Zhu, Y. (2020, January 18–22). Channel pruning via optimal thresholding. Proceedings of the 27th International Conference, ICONIP 2020, Bangkok, Thailand.10.1007/978-3-030-63823-8_58 ↗
28. Visual Tracking Algorithm Based on Online Feature Discrimination with Siamese Network10.3788/AOS201939.0915003 ↗
29. Wu, Y., Lim, J., and Yang, M. (2013, January 23–28). Online object tracking: A benchmark. Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition, Portland, OR, USA.10.1109/CVPR.2013.312 ↗
30. Wang, Q., Gao, J., Xing, J., Zhang, M., and Hu, W. (2017). Dcfnet: Discriminant correlation filters network for visual tracking. arXiv.
31. Fundamental concepts in statistics: Elucidation and illustration10.1152/jappl.1998.85.3.775 ↗
32. Danelljan, M., Hager, G., Shahbaz Khan, F., and Felsberg, M. (2015, January 7–13). Learning spatially regularized correlation filters for visual tracking. Proceedings of the 2015 IEEE International Conference on Computer Vision (ICCV), Santiago, Chile.10.1109/ICCV.2015.490 ↗
33. Danelljan, M., Robinson, A., Shahbaz Khan, F., and Felsberg, M. (2016, January 11–14). Beyond correlation filters: Learning continuous convolution operators for visual tracking. Proceedings of the 4th European Conference, Amsterdam, The Netherlands.10.1007/978-3-319-46454-1_29 ↗
34. Danelljan, M., Bhat, G., Shahbaz Khan, F., and Felsberg, M. (2017, January 21–26). Eco: Efficient convolution operators for tracking. Proceedings of the Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA.10.1109/CVPR.2017.733 ↗
35. Image quality assessment: From error visibility to structural similarity10.1109/TIP.2003.819861 ↗
36. Imagenet large scale visual recognition challenge10.1007/s11263-015-0816-y ↗