Volumetric model and 3D trajectory of a moving car derived from monocular TV frame sequences of a street scene

https://doi.org/10.1016/0146-664X(82)90081-8Get rights and content

Abstract

A polyhedral approximation for the volumetric description of a moving rigid object in a real-world scene is derived, based on measurements made on monocular TV frame sequences. The trajectory and attitude of the object motion relative to the camera is simultaneously determined up to the same factor which scales the object description. Results from two street scene sequences are presented. The approach is compared to related ones reported in the recent literature. Our experience forced us to modify the relaxation approach ofBarnard and Thompson (IEEE Trans. Pattern Anal. Mach. Intell.PAMI-2, 1980, 333–340; Technical Report 79-1, Computer Science Department, University of Minnesota, Minneapolis, 1979) in order to obtain acceptable results. These modifications are described and discussed.

Referenceas (56)

  • MarrD. et al.

    Theory of edge detection

  • NagelH.-H.

    From Digital Picture Processing to Image Analysis

  • RoachJ.W. et al.

    Determining the movement of objects from a sequence of images

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1980)
  • AsadaM. et al.

    Three Dimensional Motion Interpretation for the Sequence of Line Drawings

  • BarnardS.T. et al.

    Disparity analysis of images

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1980)
    BarnardS.T. et al.
  • BeaudetP.R.

    Rotationally Invariant Image Operators

  • BersK.H. et al.

    Object Detection in Image Sequences

  • BondeT. et al.

    Deriving a 3-D Description of a Moving Rigid Object from Monocular TV-Frame Sequences

  • ChengJ.-K. et al.

    Algorithms for Matching Relational Structures and their Application to Image Processing

    (December 1980)
  • ClocksinW.F.

    Determining the Orientation of Surfaces from Optical Flow

  • ClocksinW.F.

    The Effect of Motion Contrast on Surface Slant and Edge Detection

  • ClocksinW.F.

    Perception of surface slant and edge labels from optical flow: A computational approach

    Perception

    (1980)
  • DreschlerL.

    Ermittlung markanter Punkte auf den Bildern bewegter Objekte und Berechnung einer 3D-Beschreibung auf dieser Grundlage

  • DudaR.O. et al.

    Pattern Classification and Scene Analysis

    (1973)
  • GenneryD.B.

    Stereo-Camera Calibration

  • GerlachH.

    Digitale Bildfolgenauswertung zum Wiederfinden von Objekten in natürlicher Umgebung

  • GilbertA.L. et al.

    A Real-Time Video Tracking System Using Image Processing Techniques

  • GilbertA.L. et al.

    A real-time video tracking system

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1980)
  • GrimsonW.E.L. et al.

    A Computer Implementation of a Theory of Human Stereo Vision

  • HimmelblauD.M.

    Applied Nonlinear Programming

    (1972)
  • HirzingerG. et al.

    Automated TV Tracking of Moving Objects—The DFVLR Tracker and Related Approaches

  • HummelR.A. et al.

    On the Foundations of Relaxation Labeling Processes

  • JacobusC.J. et al.

    Motion detection and analysis of matching graphs of intermediate-level primitives

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1980)
  • JainR. et al.

    On the analysis of accumulative difference pictures from image sequences of real world scenes

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1979)
  • KitchenL. et al.

    Gray-Level Corner Detection

  • KornA. et al.

    Motion Analysis in Natural Scenes Picked up by a Moving Optical Sensor

  • KraaschR. et al.

    Automatische Dreidimensionale Beschreibung bewegter Gegenstände

  • LandzettelK. et al.

    Konzept und Realisierung eines mit Kontrastauswertung arbeitenden TV-Trackers

  • Cited by (100)

    • Invariant Gabor-based interest points detector under geometric transformation

      2014, Digital Signal Processing: A Review Journal
      Citation Excerpt :

      In this case, the ability to correspond interest points in the presence of geometric transformation is critical. Therefore, many interest points detectors have been developed to cope with this problem [15–20]. Due to limited property of the Hessian [15,16] and Harris [17] detectors which are not invariant to scale transformation, Lindeberg [18], Lowe [19], Mikolajczyk and Schmid [20] proposed methods to deal with scale transformation.

    • Detection of moving objects in airborne thermal videos

      2006, ISPRS Journal of Photogrammetry and Remote Sensing
    View all citing articles on Scopus
    View full text