Project Summary
We deal with the problem of spatially and temporally registering multiple
video sequences of a non-rigid dynamical scene. For example, registering
multiple videos of a fountain taken from different vantage points. Our approach
is not based on frame-by-frame or volume-by-volume registration. Instead,
we use the dynamic texture framework, which models the non-rigidity of the
scene with linear dynamical systems encoding both the dynamics and the appearance
of the scene. Our key contribution is to observe that a certain appearance
matrix extracted from the dynamic texture model is invariant with respect
to rigid motions of the camera, thus it can be directly used to register
the video sequences. Our framework is applicable to both synchronized videos
as well as videos that contain a temporal lag between them. The final result is a simple and flexible
method that achieves state-of the- art performance with a significant reduction
in computational complexity.
Publication
[1]
A. Ravichandran and R. Vidal
Video Registration using Dynamic Textures.
IEEE Transactions on Pattern Analysis and Machine Intelligence, January 2011.
[2]
A. Ravichandran and R. Vidal
European Conference on Computer Vision, 2008.
[3]
A. Ravichandran and R. Vidal.
International Workshop on Dynamical Vision, October 2007.
Patents
System and Method For Registering Video Sequences. Patent Publication Number US2010-0260439-A1, October 14, 2010.
Acknowledgements
This work was partially supported by startup funds from JHU, by grants ONR
N00014-05-1083, NSF CAREER 0447739, NSF EHS-0509101, and by contract
JHU APL-934652.