TY - JOUR
T1 - Epiflow-A paradigm for tracking stereo correspondences
AU - Zhang, Hongsheng
AU - Negahdaripour, Shahriar
N1 - Funding Information:
This paper describes research supported by the NOAA Ocean Exploration Program under project NA04OAR4600039.
Funding Information:
We are grateful to the reviewers for their comments that have led to a significant improvement of our manuscript. The paper describes research under project NA04OAR4600039 supported by NOAA Ocean Exploration Program. Support from the DOD SERDP for project CS-1333 enabled the collection of the ocean data used in various experiments. Views and conclusions are those of the authors, and are not to be interpreted as opinions shared and endorsements by NOAA.
PY - 2008/9
Y1 - 2008/9
N2 - We propose the epiflow as a novel framework to integrate the motion and stereo epipolar geometries for the tracking of stereo matches. For a given stereo pair, spatio-temporal correspondences in the next stereo pair are established from the solution of an optimization problem that incorporates the epiflow geometrical constraint with the brightness constancy assumption as the photometric similarity measure. Two different implementations are proposed, each utilizing the stereo RANSAC algorithm as a mechanism to gain robustness and efficiency in handling large temporal disparity ranges, identify outliers and estimate the motion of the stereo rig as complementary information. The performance of the proposed methods is demonstrated in experiments with various complex underwater images, and comparison with results from the SIFT method allows the assessment of our novel contributions.
AB - We propose the epiflow as a novel framework to integrate the motion and stereo epipolar geometries for the tracking of stereo matches. For a given stereo pair, spatio-temporal correspondences in the next stereo pair are established from the solution of an optimization problem that incorporates the epiflow geometrical constraint with the brightness constancy assumption as the photometric similarity measure. Two different implementations are proposed, each utilizing the stereo RANSAC algorithm as a mechanism to gain robustness and efficiency in handling large temporal disparity ranges, identify outliers and estimate the motion of the stereo rig as complementary information. The performance of the proposed methods is demonstrated in experiments with various complex underwater images, and comparison with results from the SIFT method allows the assessment of our novel contributions.
KW - Epipolar geometry
KW - Feature matching and tracking
KW - Motion stereo
KW - Stereo correspondences
UR - http://www.scopus.com/inward/record.url?scp=50249179308&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=50249179308&partnerID=8YFLogxK
U2 - 10.1016/j.cviu.2008.01.001
DO - 10.1016/j.cviu.2008.01.001
M3 - Article
AN - SCOPUS:50249179308
VL - 111
SP - 307
EP - 328
JO - Computer Vision and Image Understanding
JF - Computer Vision and Image Understanding
SN - 1077-3142
IS - 3
ER -