The Community for Technology Leaders
2013 IEEE Conference on Computer Vision and Pattern Recognition (2004)
Washington, D.C., USA
June 27, 2004 to July 2, 2004
ISSN: 1063-6919
ISBN: 0-7695-2158-4
pp: 746-753
Rodrigo L. Carceroni , Universidade Federal de Minas Gerais
Flávio L. C. Pádua , Universidade Federal de Minas Gerais
Kiriakos N. Kutulakos , University of Toronto
Geraldo A. M. R. Santos , Universidade Federal de Minas Gerais
We present a novel approach for temporally aligning N unsynchronized sequences of a dynamic 3D scene, captured from distinct viewpoints. Unlike existing methods, which work for N = 2 and rely on a computationally-intensive search in the space of temporal alignments, we reduce the problem for general N to the robust estimation of a single line in ℜN. This line captures all temporal relations between the sequences and can be computed without any prior knowledge of these relations. Experimental results show that our method can accurately align sequences even when they have large mis-alignments (e.g., hundreds of frames), when the problem is seemingly ambiguous (e.g., scenes with roughly periodic motion), and when accurate manual alignment is difficult (e.g., due to slow-moving objects).
Rodrigo L. Carceroni, Flávio L. C. Pádua, Kiriakos N. Kutulakos, Geraldo A. M. R. Santos, "Linear Sequence-to-Sequence Alignment", 2013 IEEE Conference on Computer Vision and Pattern Recognition, vol. 01, no. , pp. 746-753, 2004, doi:10.1109/CVPR.2004.150
99 ms
(Ver 3.3 (11022016))