ABSTRACT
A method for multi-angle face tracking based on multi-feature fusion and KCF algorithm is proposed in this paper. Firstly, Haar-like, MB-LBP and HOG features are combined to detect the face region, which is regarded as the initial tracking search window. Secondly, training, updating and adjusting search window size. Finally, if the tracking target is lost, face region is relocated through the multi-angle detection algorithm which can realize real-time and automatic tracking. To verify the effect, our method is compared with the traditional KCF algorithm. The results of simulation experiments show that, the method proposed can track multi-angle face of video sequences from video database in real time, the center distance error is 8.49, the overlap rate is 58.78%, and the tracking time is 55.41ms.
- Yuan C, Duan-Sheng C. J. 2010. A Multi-View Face Tracking Method Syncretized LBP Texture Feature. Journal of Huaqiao University(Natural Science). 31(3), 282--287.DOI= http://dx.doi.org/10.11830/ISSN.1000-5013.2010.03.0282.Google Scholar
- Rui Z, Huai-Yu W U, Ruo-Hong W U. J. 2016. Robust face tracking algorithm based on strong tracking Kalman filter. Computer Engineering & Design. 37(2), 475--480. DOI= http://dx.doi.org/10.16208/j.issn1000-7024.2016.02.037.Google Scholar
- Kalal Z, Mikolajczyk K, Matas J.J.2012 J. Tracking-Learning-Detection.IEEE Transactions on Pattern Analysis & Machine Intelligence. 34(7), 1409--22. DOI= http://dx.doi.org/10.1109/TPAMI.2011.239. Google ScholarDigital Library
- Henriques J F, Rui C, Martins P, et al. J. 2015. High-Speed Tracking with Kernelized Correlation Filters. IEEE Transactions on Pattern Analysis & Machine Intelligence. 37(3), 583--596. DOI=http://dx.doi.org/10.1109/TPAMI.2014.2345390.Google ScholarDigital Library
- ZHANGL, WANG YJ, SUN H H, et al. J. 2016. Adaptive scale object tracking with kernelized correlation filters. Optics and Precision Engineering. 24(2), 448--459. DOI=http://dx.doi.org/10.3788/OPE.20162402.0448.Google ScholarCross Ref
- Xin Li, Qiao Liu, Zhenyu He, Hongpeng Wang, Chunkai Zhang, and Wen-Sheng Chen. J.2016. A multi-view model for visual tracking via correlation filters. Know.-Based Syst. 113, C (December 2016), 88--99. DOI= https://doi.org/10.1016/j.knosys.2016.09.014 Google ScholarDigital Library
- Xu Y, Liu X, Liu Y, et al, C. 2016. Multi-view People Tracking via Hierarchical Trajectory CompositionIn. In Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)( Las Vegas, USA, June 27--30, 2016). IEEE Computer Society, Washington, DC, USA, 4256--4265. DOI= https://doi.org/10.1109/CVPR.2016.461Google ScholarCross Ref
- Hu M, Liu Y, Wang R. C. 2017. The Research of Multi-angle Face Detection Based on Multi-feature.In Proceedings of 9th International Conference on Image and Graphics (Shanghai, China, September 13-15, 2017) Springer, Singapore 466--476.DOI=https://doi.org/10.1007/978-3-319-71607-7_41Google Scholar
- Yi Wu, Jongwoo Lim, and Ming-Hsuan Yang. C. 2013. Online Object Tracking: A Benchmark. In Proceedings of the 2013 IEEE Conference on Computer Vision and Pattern Recognition (CVPR '13). (Portland, OR, USA, June 23--28, 2013). IEEE Computer Society, Washington, DC, USA, 2411--2418. DOI: https://doi.org/10.1109/CVPR.2013.312 Google ScholarDigital Library
- Joost van de Weijer, Cordelia Schmid, Jakob Verbeek, and Diane Larlus. J. 2009. Learning color names for real-world applications. Trans. Img. Proc. 18, 7 (July 2009), 1512--1523. DOI=http://dx.doi.org/10.1109/TIP.2009.2019809 Google ScholarDigital Library
- Everingham M, Gool L, Williams C K, et al. J. 2010. The Pascal Visual Object Classes (VOC) Challenge. International Journal of Computer Vision (2010), 88(2), 303--338.DOI=. https://doi.org/10.1007/s11263-009-0275-4 Google ScholarDigital Library
Index Terms
- The Research on Multi-angle Face Tracking Based on Multi-feature Fusion and KCF
Recommendations
Long Time Target Tracking Algorithm Based on Multi Feature Fusion and Correlation Filtering
AIPR '21: Proceedings of the 2021 4th International Conference on Artificial Intelligence and Pattern RecognitionThis paper considers the problem of long-term target tracking in complex scenes when tracking failures are unavoidable due to illumination change, target deformation, scale change, motion blur, and other factors. More specifically, we propose a target ...
3D Face Recognition Using Multi-level Multi-feature Fusion
PSIVT '10: Proceedings of the 2010 Fourth Pacific-Rim Symposium on Image and Video TechnologyThis paper proposed a novel 3D face recognition algorithm using multi-level multi-feature fusions. A new face representation method named average edge image is proposed in addition to traditional ones such as maximal principal curvature image and range ...
Multi-feature fusion network for facial expression recognition in the wild
Facial expression recognition (FER) has been one of the research focuses in recent years due to its significance in human-computer interactions. However, there are still challenges in the field of FER caused by the diversity and variation of facial ...
Comments