Robust object tracking via multi-feature adaptive fusion based on stability: contrast analysis

详细信息    查看全文

• 作者：Zhiyong Li ; Shuang He ; Mervat Hashem
• 关键词：Object tracking ; Feature fusion ; Multi ; feature joint descriptor ; Stability ; Contrast
• 刊名：The Visual Computer
• 出版年：2015
• 出版时间：October 2015
• 年：2015
• 卷：31
• 期：10
• 页码：1319-1337
• 全文大小：4,632 KB
• 参考文献：1.Wang, J., Bebis, G., Miller, R.: Robust video-based surveillance by integrating target detection with tracking. In: Proc. Conf. CVPRW OTCBVS, pp. 137鈥?45 (2006)
  2.Stauffer, C., Grimson, W.: Learning patterns of activity using real time tracking. IEEE Trans. Patt. Anal. Mach. Intell. 22(8), 747鈥?57 (2000)CrossRef
  3.Sun, L., Klank, U., Beetz, M.: EYEWATCHME: 3-D hand and object tracking for inside out activity analysis. In: Proc. IEEE Comput. Soc Conf. CVPR Workshops, pp. 9鈥?6 (2009)
  4.Nguyen, T.H.D., Qui, T.C.T., Xu, K.: Real-time 3D human capture system for mixed-reality art and entertainment. IEEE Trans. Vis. Comput. Gr. 11(6), 706鈥?21 (2005)CrossRef
  5.Luo, H., Ci, S., Wu, D., Stergiou, N., Siu, K.: A remote markerless human gait tracking for e-healthcare based on content-aware wireless multimedia communications. IEEE Wireless Commun. 17(1), 44鈥?0 (2010)CrossRef
  6.Hu, W., Tan, T., Wang, L., Maybank, S.: A survey on visual surveillance of object motion and behaviors. IEEE Trans. Syst. Man Cyber.-C 34(3), 334鈥?52 (2004)CrossRef
  7.Kalal, Z., Mikolajczyk, K., Matas, J.: Tracking鈥搇earning鈥揹etection. IEEE. Trans. Pattern. Anal. Mach. Intell. 34(7), 1409鈥?422 (2012)CrossRef
  8.Babenko, B., Yang, M.H., Belongie, S.: Robust object tracking with online multiple instance learning. Pattern Anal. Mach. Intell. 33(8), 1619鈥?632 (2011)CrossRef
  9.Zhang, K., Song, H.: Real-time visual tracking via online weighted multiple instance learning. Pattern Recognit. 46(1), 397鈥?11 (2013)MATH CrossRef
  10.Sun, L., Liu, G.: Visual object tracking based on combination of local description and global representation. IEEE Trans. Circuits Syst. Video Technol. 21(4), 408鈥?20 (2011)CrossRef
  11.Zoidi, O., Tefas, A., Pitas, I.: Visual object tracking based on local steering kernels and color histograms. IEEE Trans. Circuits Syst. Video Technol. 23(5), 870鈥?82 (2013)CrossRef
  12.Wu, B., Nevatia, R.: Optimizing discrimination-efficiency tradeoff in integrating heterogeneous local features for object detection. In: CVPR, pp. 1鈥? (2008)
  13.Wang, X., Han, T., Yan, S.: A HOG鈥揕BP human detector with partial occlusion handling. In: ICCV, pp. 32鈥?9 (2009)
  14.Yang, F., Lu, H., Yang, M.: Robust visual tracking via multiple kernel boosting with affinity constraints. IEEE Trans. Circuits Syst. Video Technol. 24, 242鈥?54 (2013)CrossRef
  15.Zhou, H., Yuan, Y., Shi, C.: Object tracking using SIFT features and mean shift. Comput. Vis. Image Underst. 113(3), 345鈥?52 (2009)
  16.Stern, H., Efros, B.: Adaptive color space switching for face tracking in multi-color lighting environment. In: Proc. IEEE Int. Conf. on Automatice Face and Gesture Recognition, Washington, DC, pp. 249鈥?54 (2002)
  17.Wang, J., Yagi, Y.: Integrating color and shape-texture features for adaptive real-time object tracking. IEEE Trans. Image Process. 17(2), 235鈥?40 (2008)CrossRef
  18.Nedovic, V., Liem, M., Corzilius, M., Smids, M.: Kernel-based object tracking using adaptive feature selection. Project Report (2005)
  19.Woodley, T., Stenger, B., Cipolla, R.: Tracking using online feature selection and a local generative model. BMVC, pp. 1鈥?0 (2007)
  20.Zhang, K., Zhang, L., Yang, M.: Real-time object tracking via online discriminative feature selection. IEEE Trans. Image Process. 22(12), 4664鈥?677 (2013)MathSciNet CrossRef
  21.Zhang, L., Zhang, K., Yang, M., et al.: Robust object tracking via active feature selection. IEEE Trans. Circuits Syst. Video Technol. 23(11), 1957鈥?967 (2013)CrossRef
  22.Li, G., Huang, Q., Pang, J., Jiang, S., Qin, L.: Online selection of the best k-feature subset for object tracking. Elsevier, 23(2), pp. 254鈥?63 (2011)
  23.Yoon, J.H., Kim, D.Y., Yoon, K.J.: Visual tracking via adaptive tracker selection with multiple features. Computer Vision鈥擡CCV 2012. Springer, Berlin, pp. 28鈥?1 (2012).
  24.Spengler, M., Schiele, B.: Towards robust multi-cue integration for visual tracking. Mach. Vis. Appl. 14(1), 50鈥?8 (2003)CrossRef
  25.Leichter, I., Lindenbaum, M., Rivlin, E.: A general framework for combining visual trackers鈥攖he 鈥渂lack boxes鈥?approach. Int. J. Comput. Vis. 67(3), 343鈥?63 (2006)CrossRef
  26.Badrinarayanan, V., Perez, P., Le Clerc, F., et al.: Probabilistccolor and adaptive multi-feature tracking with dynamically switched priority between cues. Computer Vision, ICCV, IEEE 11th International Conference on IEEE, pp. 1鈥? (2007)
  27.Collins, R., Liu, Y., Leordeanu, M.: Online selection of discriminative tracking features. IEEE Trans. Pattern Anal. Mach. Intell. 27(10), 1631鈥?643 (2005)CrossRef
  28.Comaniciu, D., Ramesh, V., Meer, P.: Kernel-based object tracking. IEEE Trans. Pattern Anal. Mach. Intell. 25(5), 564鈥?77 (2003)CrossRef
  29.Kailath, T.: The divergence and Bhattacharyya distance measures in signal selection. IEEE Trans. Commun. Technol. 15(1), 52鈥?0 (1967)CrossRef
  30.Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. Comput. Vis. Pattern Recognit 1, 886鈥?93 (2005)
  31.Ning, J., Zhang, L., Zhang, D., Wu, C.: Robust mean-shift tracking with corrected background-weighted histogram. IET Comput. Vis. 6, 62鈥?9 (2012)MathSciNet CrossRef
  32.Wang, Q., Chen, F., Yang, J., Xu, W., Yang, M.: Transferring visual prior for online object tracking. IEEE Trans. Image Process. 21(7), 3296鈥?305 (2012)MathSciNet CrossRef
  33.Zhang, K., Zhang, L., Yang, M.H.: Real-time compressive tracking. Computer Vision鈥擡CCV 2012. Springer, Berlin, pp. 864鈥?77 (2012)
  
• 作者单位：Zhiyong Li (1)
  Shuang He (1)
  Mervat Hashem (1)
  
  1. College of Computer Science and Electronic Engineering, Hunan University, Changsha, 410082, China
  
• 刊物类别：Computer Science
• 刊物主题：Computer Graphics
  Computer Science, general
  Artificial Intelligence and Robotics
  Image Processing and Computer Vision
  
• 出版者：Springer Berlin / Heidelberg
• ISSN：1432-2315

文摘

Object tracking under complex circumstances is a challenging task because of background interference, obstacle occlusion, object deformation, etc. Given such conditions, robustly detecting, locating, and analyzing a target through single-feature representation are difficult tasks. Global features, such as color, are widely used in tracking, but may cause the object to drift under complex circumstances. Local features, such as HOG and SIFT, can precisely represent rigid targets, but these features lack the robustness of an object in motion. An effective method is adaptive fusion of multiple features in representing targets. The process of adaptively fusing different features is the key to robust object tracking. This study uses a multi-feature joint descriptor (MFJD) and the distance between joint histograms to measure the similarity between a target and its candidate patches. Color and HOG features are fused as the tracked object of the joint representation. This study also proposes a self-adaptive multi-feature fusion strategy that can adaptively adjust the joint weight of the fused features based on their stability and contrast measure scores. The mean shift process is adopted as the object tracking framework with multi-feature representation. The experimental results demonstrate that the proposed MFJD tracking method effectively handles background clutter, partial occlusion by obstacles, scale changes, and deformations. The novel method performs better than several state-of-the-art methods in real surveillance scenarios. Keywords Object tracking Feature fusion Multi-feature joint descriptor Stability Contrast