利用密度描述符对应的视觉跟踪算法

查宇飞 1; 杨源 2; 王锦江 3; 张立朝 1

doi:10.7652/xjtuxb201409003

您当前的位置：

首页 >

文章列表页 >

利用密度描述符对应的视觉跟踪算法

更新时间：2025-07-09

- 利用密度描述符对应的视觉跟踪算法
- A Visual Object Tracking Algorithm Using Dense Descriptors Correspondences
- 西安交通大学学报 2014年48卷第9期页码：13-18
- 作者机构：
  
  空军工程大学航空航天工程学院,西安,710038
- 作者简介：
- 基金信息：
  
  : 2014-04-01。作者简介: 查宇飞(1979—),男,博士,副教授。基金项目: 国家自然科学基金资助项目(6120 2339,61203628);陕西省自然科学基金资助项目(2012JQ8034)
- DOI：10.7652/xjtuxb201409003
  中图分类号： TP391
- 网络首发：2014-09-10，
  
  纸质出版：2014
- 稿件说明：
移动端阅览
查宇飞 1, 杨源 2, 王锦江 3, 等. 利用密度描述符对应的视觉跟踪算法[J]. 西安交通大学学报, 2014,48(9):13-18.

A Visual Object Tracking Algorithm Using Dense Descriptors Correspondences[J]. 2014, 48(9): 13-18.
查宇飞 1, 杨源 2, 王锦江 3, 等. 利用密度描述符对应的视觉跟踪算法[J]. 西安交通大学学报, 2014,48(9):13-18. DOI： 10.7652/xjtuxb201409003.

A Visual Object Tracking Algorithm Using Dense Descriptors Correspondences[J]. 2014, 48(9): 13-18. DOI： 10.7652/xjtuxb201409003.

摘要

针对稀疏描述符由于关键点检测不稳定及其误匹配造成跟踪失败的问题

提出了一种利用密度描述符对应实现目标跟踪的算法。该算法通过计算目标在相邻两帧之间的密度描述符流

考虑目标的空间分布特性和描述符的权重

得到目标的运动矢量

获得目标在当前帧中的估计; 根据密度描述符的运动矢量与目标运动矢量的关系及其匹配程度

更新密度描述符的权重。在大量测试数据上对所提算法进行实验

并与相关跟踪算法进行比较

定性和定量分析表明:当发生光照、遮挡、姿态变化时

所提算法能够稳定地跟踪目标

跟踪成功率均在90%以上

跟踪误差要小于其他算法。

Abstract

An object tracking algorithm is proposed to overcome the problem that the sparse local invariance feature descriptor depends on the feature detection and always leads to failure. The proposed algorithm uses the dense descriptors correspondences. The dense descriptor flows of objects between consecutive frames are calculated

and object estimations are obtained by considering the spatial distribution and weights of dense flows. Then the weights of descriptors are updated according to the relation and the matching degree between the descriptor motion and the object motion. Qualitative and quantitative analyses on challenging benchmark image sequences show that the average tracking successes rate of the proposed algorithm is over 90% and the performance of the algorithm is better than the performances of several state-of-art methods that assume the constant brightness and template.

关键词

Keywords

references

SENST T, EISELEIN V, SIKORA T. Robust local optical flow for feature tracking[J]. IEEE Transactions on Circuits and Systems, 2012, 22(9): 1377-1387.

LEE K Y, PARK R H, LEE S W. Color matching for soft proofing using a camera[J]. IET Image Process, 2012, 6(3): 292-300.

PEREZ P, HUE C, VERMAAK J, et al. Color-based probabilistic tracking[C]∥Proceedings of 7th European Conference on Computer Vision. Berlin, Germany: Springer, 2002: 661-675.

COMANICIU D, RAMESH V, MEER P. Kernel-based object tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2003, 25(5): 564-577.

ROSS D, LIM J, LIN Ruei-Sung, et al. Incremental learning for robust visual tracking[J]. International Journal of Computer Vision, 2007, 77(1): 125-141.

DAVID G L. Scale affine invariant interest point detectors[J]. International Journal of Computer Vision, 2004, 60(1): 63-86.

CALONDER M, LEPETIT V, FUA P. Binary robust independent elementary features[C]∥Proceedings of 11th European Conference on Computer Vision. Berlin, Germany: Springer, 2010: 778-792.

HEINLY J, DUMN E, FRAHM J M. Comparative evaluation of binary features[C]∥Proceedings of 12th European Conference on Computer Vision. Berlin, Germany: Springer, 2012: 369-382.

MIKOLAJCZYK K, SCHMID C. A performance evaluation of local descriptors[J]. IEEE Trans on Pattern Analysis and Machine Intelligence, 2005, 27(10): 1615-1630.

ZHOU Huiyu, YUAN Yuan, SHI Chunmei. Object tracking using SIFT features and mean shift[J]. Computer Vision and Image Understand, 2009, 113(3): 345-352.

ENGIN T, VINCENT L, PASCAL F. An efficient dense descriptor applied to wide baseline stereo[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 32(5): 815-830

VEDALDI A, FULKERSON B. An open and portable library of computer vision algorithms[C]∥Proceedings of the International Conference on Multimedia. New York, USA: ACM, 2010: 1469-1472

LIU Ce, JENNY Y, ANTONIO T, et al. Dense correspondence across different scenes[C]∥Proceedings of 9th European Conference on Computer Vision. Berlin, Germany: Springer, 2008: 28-42.

BAY H, TUYTELAARS T, GOOL L V. Speeded up robust features[C]∥Proceedings of 7th European Conference on Computer Vision. Berlin, Germany: Springer, 2006: 404-417

BARNES C, SHECHTMAN E, GOLDMAN D, et al. The generalized PatchMatch correspondence algorithm[C]∥Proceedings of 11th European Conference on Computer Vision. Berlin, Germany: Springer, 2010: 29-43.

KIM J, LIU Ce, SHA Fei, et al. Deformable spatial pyramid matching for fast dense correspondences[C]∥26th IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2013: 2307-2314.

BABENKO B, YANG Ming-Hsuan, BELONGIE S. Visual tracking with online multiple instance learning[C]∥24th IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2009: 983-990.

SANTNER J, LEISTNER C, SAFFARI A, et al. Parallel robust online simple tracking[C]∥ Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. Piscataway, NJ, USA: IEEE, 2010: 723-730.

KALAL Z, MIKOLAJCZYK K, MATAS J. Tracking learning detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2010, 6(1): 1-14

蔺海峰, 马宇峰, 宋涛. 基于SIFT特征目标跟踪算法研究[J]. 自动化学报, 2010, 36(8): 1204-1208.

LIN Haifeng, MA Yufeng, SONG Tao. Research on object tracking algorithm based on SIFT[J]. Acta Automatica Sinica, 2010, 36(8): 1204-1208.

ZULIANI M, KENNEY C S, MANJUNATH B S. The multiRANSAC algorithm and its application to detect planar homographies[C]∥Proceedings IEEE International Conference on Image Processing. Piscataway, NJ, USA: IEEE, 2005: 153-156.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

快速多特征金字塔的尺度目标跟踪方法

目标跟踪中基于深度可分离卷积的剪枝方法

结构-语义-残差协同的图像压缩框架：面向隐私保护与高保真重建

超临界二氧化碳向心透平轮背间隙内风阻损失及流动特性研究

采用试验设计方法的第三代CO2气体箔轴承箔片几何参数敏感性分析