基于区域预估与自适应分类的视觉跟踪算法

针对视觉跟踪中的目标形变、部分遮挡和平面外旋转等问题,提出一种基于区域预估与自适应分类的视觉跟踪算法。该方法基于跟踪-修正-检测框架,利用Mean-Shift算法进行跟踪,并使跟踪器与检测器紧密相连,利用修正模块判断跟踪器和检测器是否需要在线更新;采用Kalman滤波器对目标潜在位置区域进行预估,避免全局扫描的繁琐流程;所提出的自适应方差分类器能够动态地调整分类器参数,增强分类器的灵活性,提高跟踪稳健性。采用OTB-2013评估基准中的视频序列进行测试,并将所提算法与其他4种具有代表性的视觉跟踪算法进行对比,实验结果表明,所提算法的稳健性和准确性均优于对比算法。

Abstract

Aim

ing at the problems of target deformation, partial occlusion, and out-of-plane rotation in visual tracking, we propose a visual tracking method based on region estimation and adaptive classification. The method is based on tracking-learning-detection framework. Firstly, we use the Mean-Shift algorithm to realize the tracking, and the tracker is closely connected with the detector. The correction module determines whether the detector needs to be updated online or not. Secondly, we use the Kalman filter to estimate the potential location of the target in order to avoid cumbersome global scanning. Finally, the proposed adaptive variance classifier dynamically adjusts the classifier parameters, enhances the flexibility of the classifier, and improves robustness. Experiments perform on the OTB-2013 evaluation benchmark show that the robustness and accuracy of the proposed algorithm are better than those of contrastive algorithms.

参考文献

[1] . Object tracking: a survey[J]. ACM Computing Surveys, 2006, 38(4): 1-45.

. Scale adaptive kernel correlation filtering for target tracking[J]. Laser & Optoelectronics Progress, 2018, 55(4): 041501.

, et al. Infrared dim-small target tracking based on guided image filtering and kernelized correlation filtering[J]. Acta Optica Sinica, 2018, 38(2): 0204004.

[4] Bao CL, WuY, Ling HB, et al. Real time robust L1 tracker using accelerated proximal gradient approach[C]∥2012 IEEE Conference on Computer Vision and Pattern Recognition, June 16-21, 2012, Providence, RI,USA. New York: IEEE, 2012: 1830- 1837.

[5] , et al. Robust visual tracking via basis matching[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2017, 27(3): 421-430.

[6] , 等. 基于HSV颜色特征和贡献度重构的行人跟踪[J]. 激光与光电子学进展, 2017, 54(9): 091004.

, et al. Pedestrian tracking based on HSV color features and reconstruction by contributions[J]. Laser & Optoelectronics Progress, 2017, 54(9): 091004.

[7] BabenkoB, Yang MH, BelongieS. Visual tracking with online multiple instance learning[C]∥2009 IEEE Conference on Computer Vision and Pattern Recognition, June 20-25, 2009, Miami, FL, USA. New York: IEEE, 2009: 983- 990.

[8] , et al. Struck: structured output tracking with kernels[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(10): 2096-2109.

[9] . Tracking-learning-detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(7): 1409-1422.

[10] MaC, Yang XK, Zhang CY, et al. Long-term correlation tracking[C]∥2015 IEEE Conference on Computer Vision and Pattern Recognition, June 7-12, 2015, Boston, MA, USA. New York: IEEE, 2015: 5388- 5396.

[11] HuangC, LuceyS, RamananD. Learning policies for adaptive tracking with deep feature cascades[C]∥2017 IEEE International Conference on Computer Vision (ICCV), October 22-29, 2017, Venice, Italy.New York: IEEE, 2017, 1: 105- 114.

[12] HeldD, ThrunS, SavareseS. Learning to track at 100 FPS with deep regression networks[M] ∥Leibe B, Matas J, Sebe N, et al.Computer vision-ECCV 2016. Lecture notes in computer science. Cham: Springer, 2016, 9905: 749- 765.

[13] Song YB, MaC, Wu XH, et al. VITAL: visual tracking via adversarial learning[C]∥2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, June 18-23, 2018, Salt Lake City, UT, USA. New York: IEEE, 2018: 8990- 8999.

[14] . A new approach to linear filtering and prediction problems[J]. Journal of Basic Engineering, 1960, 82(1): 35-45.

[15] KalalZ, MatasJ, MikolajczykK. P-N learning: bootstrapping binary classifiers by structural constraints[C]∥2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, June 13-18, 2010, San Francisco, CA, USA. New York: IEEE, 2010: 49- 56.

[16] . 基于自适应混合滤波的多目标跟踪算法[J]. 光学学报, 2010, 30(9): 2554-2561.

. Multi-object tracking algorithm based on adaptive mixed filtering[J]. Acta Optica Sinica, 2010, 30(9): 2554-2561.

[17] . Mean shift: a robust approach toward feature space analysis[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5): 603-619.

[18] . 检测区域自适应调整的TLD多目标跟踪算法[J]. 东北大学学报(自然科学版), 2017, 38(2): 214-218.

. A multiple-object tracking algorithm using TLD-based adaptive adjustment of detection areas[J]. Journal of Northeastern University(Natural Science), 2017, 38(2): 214-218.

[19] ViolaP, JonesM. Rapid object detection using a boosted cascade of simple features[C]∥2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, December 8-14, 2001, Kauai, USA. New York: IEEE, 2001: I511- I518.

[20] , et al. Context and profile based cascade classifier for efficient people detection and safety care system[J]. Multimedia Tools and Applications, 2013, 63(1): 27-44.

[21] Barba-GuamanL, Quezada-Sarmiento P A, Calderon-Cordova C, et al. Detection of the characters from the license plates by cascade classifiers method[C]∥2016 Future Technologies Conference (FTC), December 6-7, 2016, San Francisco, CA, USA. New York: IEEE, 2016: 560- 566.

[22] WuY, LimJ, Yang MH. Online object tracking: a benchmark[C]∥2013 IEEE Conference on Computer Vision and Pattern Recognition, June 23-28, 2013, Portland, OR, USA. New York: IEEE, 2013: 2411- 2418.

[23] Henriques JF, CaseiroR, MartinsP, et al. Exploiting the circulant structure of tracking-by-detection with kernels[C]∥2012 European conference on computer vision, October 7-13, Firenze, Italy, 2012: 702- 715.

孙彦景, 张丽颖, 云霄. 基于区域预估与自适应分类的视觉跟踪算法[J]. 激光与光电子学进展, 2019, 56(18): 181001. 孙彦景, 张丽颖, 云霄. Visual Tracking Algorithm Based on Region Estimation and Adaptive Classification[J]. Laser & Optoelectronics Progress, 2019, 56(18): 181001.

基于区域预估与自适应分类的视觉跟踪算法

关于本站 Cookie 的使用提示

全站搜索

基于区域预估与自适应分类的视觉跟踪算法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索