TY - GEN
T1 - Motion and structure factorization and segmentation of long multiple motion image sequences
AU - Debrunner, Chris
AU - Ahuja, Narendra
N1 - Funding Information:
This paper makes use of an algorithm for factorization of a measurement matrix into separate motion and structure matrices as reported by the authors in [DA1]. Subsequently in [TK1], To-masi and Kanade present a similar factorization-based method which allows arbitrary rotations, but does not have the capability to process trajectories starting and ending at arbitrary frames. Furthermore, it appears that some assumptions about the magnitude or smoothness of motion are * Supported by DARPA and the NSF under grant IRI-89-02728, and the State of Illinois Departraent of Commerce and Community Affairs under grant 90-103.
Publisher Copyright:
© Springer-Verlag Berlin Heidelberg 1992.
PY - 1992
Y1 - 1992
N2 - This paper presents a computer algorithm which, given a dense temporal sequence of intensity images of multiple moving objects, will separate the images into regions showing distinct objects, and for those objects which are rotating, will calculate the three-dimensional structure and motion. The method integrates the segmentation of trajectories into subsets corresponding to different objects with the determination of the motion and structure of the objects. Trajectories are partitioned into groups corresponding to the different objects by fitting the trajectories from each group to a hierarchy of increasingly complex motion models. This grouping algorithm uses an efficient motion estimation algorithm based on the factorization of a measurement matrix into motion and structure components. Experiments are reported using two real image sequences of 50 frames each to test the algorithm.
AB - This paper presents a computer algorithm which, given a dense temporal sequence of intensity images of multiple moving objects, will separate the images into regions showing distinct objects, and for those objects which are rotating, will calculate the three-dimensional structure and motion. The method integrates the segmentation of trajectories into subsets corresponding to different objects with the determination of the motion and structure of the objects. Trajectories are partitioned into groups corresponding to the different objects by fitting the trajectories from each group to a hierarchy of increasingly complex motion models. This grouping algorithm uses an efficient motion estimation algorithm based on the factorization of a measurement matrix into motion and structure components. Experiments are reported using two real image sequences of 50 frames each to test the algorithm.
UR - http://www.scopus.com/inward/record.url?scp=30244450166&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=30244450166&partnerID=8YFLogxK
U2 - 10.1007/3-540-55426-2_24
DO - 10.1007/3-540-55426-2_24
M3 - Conference contribution
AN - SCOPUS:30244450166
SN - 9783540554264
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 217
EP - 221
BT - Computer Vision - ECCV 1992 - 2nd European Conference on Computer Vision, Proceedings
A2 - Sandini, Giulio
PB - Springer
T2 - 2nd European Conference on Computer Vision, ECCV 1992
Y2 - 19 May 1992 through 22 May 1992
ER -