Matlab Code for Model-Based Numerical Pose Recovery
The Matlab code provided here was used in the empirical evaluation of our
fully projective reformulation of Lowe's pose recovery algorithm.
A detailed description of the algorithms implemented here can be found at:
H. Araújo, R. L. Carceroni and C. M. Brown,
``A Fully Projective Formulation to Improve the Accuracy of Lowe's
Pose-Estimation Algorithm'', in
Computer Vision and Image Understanding, to appear.
Size: 16 pages, available
for download as a gzipped PostScript
file.
H. Araújo, R. L. Carceroni and C. M. Brown, ``A Fully Projective
Formulation for Lowe's Tracking Algorithm'', TR 641, Comp. Sci. Dept.,
U. Rochester, Nov 1996.
Size: 42 pages, available for
download
as a gzipped PostScript file.