Online Boosting Tracking with Fragmented Model

详细信息    查看全文

• 作者：Dingcheng Shen (22) (23)
  Hua Zhang (22) (23)
  Yanbing Xue (22)
  Guangping Xu (22) (23)
  Zan Gao (22) (23)
  
• 关键词：on ; line boosting ; fragment ; voting weight ; drift
• 刊名：Lecture Notes in Computer Science
• 出版年：2013
• 出版时间：2013
• 年：2013
• 卷：7733
• 期：1
• 页码：521-524
• 全文大小：360KB
• 参考文献：1. Collins, R., Liu, Y., Leordeanu, M.: Online selection of discriminative tracking features. IEEE Trans. PAMI聽27(10), 1631鈥?643 (2005) CrossRef
  2. Lim, J., Ross, D., Lin, R., Yang, M.: Incremental learning for visual tracking. In: Saul, L.K., Weiss, Y., Bottou, L. (eds.) NIPS, vol.聽(17), pp. 793鈥?00. MIT Press (2005)
  3. Avidan, S.: Ensemble tracking. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp. 494鈥?01 (2005)
  4. Grabner, H., Bischof, H.: On-line boosting and vision. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp. 260鈥?67 (2006)
  5. Grabner, H., Grabner, M., Bischof, H.: Real-time tracking via on-line boosting. In: Proceedings of BMVC, vol.聽1, pp. 47鈥?6 (2006)
  6. Grabner, H., Leistner, C., Bischof, H.: Semi-supervised On-Line Boosting for Robust Tracking. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008, Part I. LNCS, vol.聽5302, pp. 234鈥?47. Springer, Heidelberg (2008) CrossRef
  7. Kalal, Z., Matas, J., Mikolajczyk, K.: P-n learning: Bootstrapping binary classifiers by structural constraints. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp. 49鈥?6 (2010)
  8. Adam, A., Rivlin, E., Shimshoni, I.: Robust fragments-based tracking using the integral histogram. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, pp. 798鈥?05 (2006)
  
• 作者单位：Dingcheng Shen (22) (23)
  Hua Zhang (22) (23)
  Yanbing Xue (22)
  Guangping Xu (22) (23)
  Zan Gao (22) (23)
  
  22. Key Laboratory of Computer Vision and System, Ministry of Education, Tianjin University of Technology, 300384, Tianjin, China
  23. Tianjin Key Laboratory of Intelligence Computing and Novel Software Technology, Tianjin University of Technology, 300384, Tianjin, China
  
• ISSN：1611-3349

文摘

We propose a novel method combining online boosting and fragment to overcome the drifting problem in on-line boosting tracking. We find that in previous on-line boosting method, the voting weights of the first few selectors are so big that the remainders can not affect the final strong classifier. This problem occurs because the voting weight of selectors are passing globally to adapt to the object variation, but usually only parts of object changes significantly in short time, and the changing part only affect its neighborhood, not the whole target area. So we divide the selector into fragments to get spatial information. The best weak classifier in each selector is combined linearly to get the final strong classifier and then find the location of the object in next frame. Experiments show robustness and generality of the proposed method.