Complete scene structure from four point correspondences
- 19 November 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 330-337
- https://doi.org/10.1109/iccv.1995.466921
Abstract
A technique is presented for computing 3D scene structure from point and line features in monocular image sequences. Unlike previous methods, the technique guarantees the completeness of the recovered scene, ensuring that every scene feature that is detected in each image is reconstructed. The approach relies on the presence of four or more reference features whose correspondences are known in all the images. Under an orthographic or affine camera model, the parallax of the reference features provides constraints that simplify the recovery of the rest of the visible scene. An efficient recursive algorithm is described that uses a unified framework for point and line features. The algorithm integrates the tasks of feature correspondence and structure recovery, ensuring that all reconstructible features are tracked. In addition, the algorithm is immune to outliers and feature drift, two weaknesses of existing structure from motion techniques. Experimental results are presented for real images.Keywords
This publication has 10 references indexed in Scilit:
- A Sequential Factorization Method for Recovering Shape and Motion from Image StreamsPublished by Defense Technical Information Center (DTIC) ,1994
- Motion from point matches using affine epipolar geometryPublished by Springer Nature ,1994
- Global surface reconstruction by purposive control of observer motionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1994
- A review of statistical data association techniques for motion correspondenceInternational Journal of Computer Vision, 1993
- Shape and motion from image streams under orthography: a factorization methodInternational Journal of Computer Vision, 1992
- Affine structure from motionJournal of the Optical Society of America A, 1991
- Trinocular stereo vision for roboticsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1991
- Finding point correspondence and determining motion of a rigid object from two weak perspective viewsComputer Vision, Graphics, and Image Processing, 1990
- Distance transformations in arbitrary dimensionsComputer Vision, Graphics, and Image Processing, 1984
- A computer algorithm for reconstructing a scene from two projectionsNature, 1981