Results are reported from a research project on three-dimensional motion analysis. The work is based on stereo image sequences of a military vehicle. The analysis depends on the segmentation of the image and the matching of features between the two images. Preliminary results are given for an extension of the work to object recognition. Extracted features and the calculated parameters of motion are incorporated into a symbolic matching system. Object recognition is performed by the symbolic inexact matching of a representation of the photographs with a database of prototypes. The use of control points to furnish ground truth is discussed.


    Access

    Access via TIB

    Check availability in my library


    Export, share and cite



    A Kalman Filter in Motion Analysis from Stereo Image Sequences

    Pan, J. N. / Shi, Y. Q. / Shu, C. Q. et al. | British Library Conference Proceedings | 1994


    A Kalman filter in motion analysis from stereo image sequences

    Pan, J.N. / Shi, Y.Q. / Shu, C.Q. | IEEE | 1994