Elsevier

Image and Vision Computing

Volume 11, Issue 7, September 1993, Pages 419-439
Image and Vision Computing

Variational approach to optical flow estimation managing discontinuities

https://doi.org/10.1016/0262-8856(93)90046-JGet rights and content

Abstract

Projection of the 3D velocity of real objects on the image plane is often called the ‘velocity field’. The estimation of this field is one of the most important research topics in computer vision. In the literature, there are numerous solutions which adopt a sort of continuity equation called optical flow constrain (OFC). The solution of this constraint equation is usually called the ‘optical flow’ field, and can be considered equal to the velocity field under particular assumptions. The structure of the OFC equation makes the optical flow estimation an ill-posed problem, like many other inverse problems in early vision. For this reason, many reglarization techniques were used in the past for estimating optical flow. The major drawback of these solutions is the presence of propagation effects which produce the loss of the information associated to the discontinuities. On the other hand, the dicontinuities are very important for estimating precise optical flow fields, and detecting the shape of moving objects. In this paper, we propose a new solution based on variational techniques for optical flow estimation and regularization, which takes into account the discontinuities, and strongly reduces the related problems. The proposed method is called ‘discontinuity-dependent variational solution ’.

References (44)

  • G.L. Fennema et al.

    Velocity determination in scene containing several moving objects

    Comput. Graph. & Image Process.

    (1979)
  • R.M. Haralick et al.

    The facet approach to optical flow

  • O Tretiak et al.

    Velocity estimation from image sequences with second order differential operators

  • A Verri et al.

    Motion field and optical flow: Qualitative properties

    IEEE Trans. PAMI

    (May 1989)
  • N.H. Nagel

    On a constraint equation for the estimation of displacement rates in image sequences

    IEEE Trans. PAMI.

    (January 1989)
  • P Nesi et al.

    A unified approach to optical flow constraint analysis

  • J Hadamard

    Sur les problems aux derivees patielles et leur signification physique

    Princeton University Bulletin

    (1902)
  • J Hadamard

    Lectures on the Cauchy problem in Linear Partial Differential Equations

    (1923)
  • A Verri et al.

    Differential techniques for optical flow

    J. Opt. Soc. Am. A

    (May 1990)
  • M Bertero et al.

    Ill-posed problems in early vision

  • F John

    Countinuous dependence on data for solutions of partial differential equations with a prescribed bound

    Commun. Pure & Appl. Math.

    (1960)
  • E DeGiorgi

    Γ-convergenza e G-convergenza

  • Cited by (62)

    • Optical flow estimation for motion-compensated compression

      2013, Image and Vision Computing
      Citation Excerpt :

      To solve the under-constrained problem, several constraints on the displacement field, such as smoothness and other assumptions, have been proposed. Typical smoothness assumptions include Horn and Schunck's regularization constraint [3], a uniform velocity assumption in a block (or template) (Lucas and Kanade [4,5]; Shi and Tomasi [16]), an intensity-gradient conservation constraint (Nagel et al. [12–14]; Nesi [15]), modeling the motion field as a Markovian random field (Konrad and Dubois [20]), and velocity field modeling with bilinear or B-splines functions (Chen et al. [28,29]). Traditionally, scientists have focused their efforts on extending the many other different methods of estimating the optical flow [1–35].

    • High-speed target tracking by fuzzy hostility-induced segmentation of optical flow field

      2009, Applied Soft Computing Journal
      Citation Excerpt :

      In these iterative methods, the velocity is evaluated at every point in the image. Regularization-based approaches using the OFC can be found in [9,31,32]. The uniqueness of these methods is that they yield the optical flow fields both inside and on the contours of the moving objects.

    • Knowledge modeling and management for mobility and transport applications

      2018, Proceedings - 4th IEEE International Conference on Collaboration and Internet Computing, CIC 2018
    • Real-Time traffic estimation of unmonitored roads

      2018, Proceedings - IEEE 16th International Conference on Dependable, Autonomic and Secure Computing, IEEE 16th International Conference on Pervasive Intelligence and Computing, IEEE 4th International Conference on Big Data Intelligence and Computing and IEEE 3rd Cyber Science and Technology Congress, DASC-PICom-DataCom-CyberSciTec 2018
    View all citing articles on Scopus
    View full text