Pose optimization based on integral of the distance between line segments

详细信息    查看全文

• 作者：YueQiang Zhang ; Xin Li ; HaiBo Liu ; Yang Shang…
• 关键词：machine vision ; perspective ; n ; line problem ; line distance function ; pose optimization ; M ; estimation
• 刊名：SCIENCE CHINA Technological Sciences
• 出版年：2016
• 出版时间：January 2016
• 年：2016
• 卷：59
• 期：1
• 页码：135-148
• 全文大小：3,645 KB
• 参考文献：1.Ferraz L, Binefa X, Moreno-Noguer F. Very fast solution to the pnp problem with algebraic outlier rejection. In: IEEE Conference on Computer Vision and Pattern Recognition, Columbus, 2014. 501–508
  2.Li S, Xu C, Xie M. A robust o (n) solution to the perspective-n-point problem. IEEE Trans Pattern Anal Mach Intell, 2012, 34: 1444–1450CrossRef
  3.Lepetit V, Moreno-Noguer F, Fua P. Epnp: An accurate o (n) solution to the pnp problem. Int J Comput Vision, 2009, 81: 155–166CrossRef
  4.Lu C, Hager G D, Mjolsness E. Fast and globally convergent pose estimation from video images. IEEE T Pattern Anal Mach Intell, 2000, 6: 610–622CrossRef
  5.Li X, Long G C, Guo P Y, et al. Accurate mirror-based camera pose estimation with explicit geometric meanings. Sci China Tech Sci, 2014, 57: 2504–2513CrossRef
  6.Dhome M, Richetin M, Lapreste J, et al. Determination of the attitude of 3D objects from a single perspective view. IEEE T Pattern Anal Mach Intell, 1989, 11: 1265–1278CrossRef
  7.Chen H H. Pose determination from line-to-plane correspondences: Existence condition and closed-form solutions. In: IEEE International Conference on Computer Vision, Osaka, 1990. 374–378
  8.Navab N, Faugeras O. Monocular pose determination from lines: Critical sets and maximum number of solutions. In: IEEE Conference on Computer Vision and Pattern Recognition, New York, 1993. 254–260CrossRef
  9.Liu Y, Huang T S, Faugeras O D. Determination of camera location from 2-D to 3-D line and point correspondences. IEEE T Pattern Anal Mach Intell, 1990, 12: 28–37CrossRef
  10.Adnan A. Linear pose estimation from points or lines. IEEE T Pattern Anal Mach Intell, 2003, 25: 578–589CrossRef
  11.Mirzaei F M, Roumeliotis S I. Globally optimal pose estimation from line correspondences. In: IEEE International Conference on Robotics and Automation, Shanghai, 2011. 5581–5588
  12.Elqursh A, Elgammal A. Line-based relative pose estimation. In: IEEE International Conference on Computer Vision, Providence, 2011. 3049–3056
  13.Zhang L, Xu C, Lee K, et al. Robust and efficient pose estimation from line correspondences. In: Asian Conference on Computer Vision, Daejeon, 2013. 217–230
  14.Liu J, Zhang X, Liu H, et al. New method for camera pose estimation based on line correspondence. Sci China Tech Sci, 2013, 56: 2787–2797CrossRef
  15.Phong Q T, Horaud R, Yassine A. Object pose from 2-D to 3-D point and line correspondences. Int J Comput Vision, 1995, 15: 225–243CrossRef
  16.Kumar R, Hanson A R. Robust methods for estimating pose and a sensitivity analysis. Cvgip Image Understand, 1994, 60: 313–342CrossRef
  17.Christy S, Horaud R. Fast and reliable object pose estimation from line correspondences. In: Computer Analysis of Images and Patterns, Kiel, 1997. 432–439CrossRef
  18.Horaud R, Dornaika F, Lamiroy B. Object pose: The link between weak perspective, paraperspective, and full perspective. Int J Comput Vision, 1997, 22: 173–189CrossRef
  19.Hanek R, Navab N, Appel M. Yet another method for pose estimation: A probabilistic approach using points lines and cylinders. In: IEEE Conference on Computer Vision and Pattern Recognition, Fort Collins, 1999. 544–550
  20.David P, Dementhon D, Duraiswami R, et al. Simultaneous pose and correspondence determination using line features. In: IEEE Conference on Computer Vision and Pattern Recognition, Madison, 2003. 424–431
  21.David P, Dementhon D, Duraiswami R, et al. SoftPOSIT: Simultaneous pose and correspondence determination. In: European Conference on Computer Vision, Copenhagen, 2002. 698–714
  22.Zhang X, Zhang Z, Li Y, et al. Robust camera pose estimation from unknown or known line correspondences. Appl Optics, 2012, 51: 936–948CrossRef
  23.Christy S, Horaud R. Iterative pose computation from line correspondences. Comput Vis Image Understand, 1999, 73: 137–144CrossRef MATH
  24.Taylor C J, Kriegman D J. Structure and motion from line segments in multiple images. IEEE T Pattern Anal Mach Intell, 1995, 17: 1021–1032CrossRef
  25.Drummond T, Cipolla R. Real-time visual tracking of complex structures. IEEE T Pattern Anal Mach Intell, 2002, 24: 932–946CrossRef
  26.Von Gioi Rafael G, Jérémie J, Morel J, et al. LSD: A fast line segment detector with a false detection control. IEEE T Pattern Anal Mach Intell, 2010, 32: 722–732CrossRef
  
• 作者单位：YueQiang Zhang (1) (2)
  Xin Li (1) (2)
  HaiBo Liu (1) (2)
  Yang Shang (1) (2)
  QiFeng Yu (1) (2)
  
  1. College of Aerospace Science and Engineering, National University of Defense Technology, Changsha, 410073, China
  2. Hunan Provincial Key Laboratory of Image Measurement and Vision Navigation, Changsha, 410073, China
  
• 刊物类别：Engineering
• 刊物主题：Chinese Library of Science
  Engineering, general
  
• 出版者：Science China Press, co-published with Springer
• ISSN：1869-1900

文摘

In this paper, new solutions for the problem of pose estimation from correspondences between 3D model lines and 2D image lines are proposed. Traditional line-based pose estimation methods rely on the assumption that the noises (perpendicular to the line) for the two endpoints are statistically independent. However, these two noises are in fact negatively correlated when the image line segment is fitted using the least-squares technique. Therefore, we design a new error function expressed by the average integral of the distance between line segments. Three least-squares techniques that optimize both the rotation and translation simultaneously are proposed in which the new error function is exploited. In addition, Lie group formalism is utilized to describe the pose parameters, and then, the optimization problem can be solved by means of a simple iterative least squares method. To enhance the robustness to outliers existing in the match data, an M-estimation method is developed to convert the pose optimization problem into an iterative reweighted least squares problem. The proposed methods are validated through experiments using both synthetic and real-world data. The experimental results show that the proposed methods yield a clearly higher precision than the traditional methods.