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.<>


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Complete scene structure from four point correspondences


    Contributors:
    Seitz, S.M. (author) / Dyer, C.R. (author)


    Publication date :

    1995-01-01


    Size :

    960781 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Complete Scene Structure from Four Point Correspondences

    Seitz, S. / Dyer, C. / IEEE Computer Society et al. | British Library Conference Proceedings | 1995


    3D Structure Reconstruction from Point Correspondences between Two Perspective Projections

    Kara, A. / Wilkes, D. M. / Kawamura, K. | British Library Online Contents | 1994


    Finding Point Correspondences in Motion Sequences Preserving Affine Structure

    Sudhir, G. / Banerjee, S. / Zisserman, A. | British Library Online Contents | 1997


    An Iterative Algorithm for Finding Point Correspondences

    Peng, Jian-Xun / Niblock, James / McMenemy, Karen | IEEE | 2008


    Radon-based structure from motion without correspondences

    Makadia, A. / Geyer, C. / Sastry, S. et al. | IEEE | 2005