| | |
Summary: To appear in the ACM SIGGRAPH 2005 conference proceedings
Panoramic Video Textures
Aseem Agarwala1 Ke Colin Zheng1 Chris Pal3 Maneesh Agrawala2 Michael Cohen2
Brian Curless1 David Salesin1,2 Richard Szeliski2
1University of Washington 2Microsoft Research 3University of Massachusetts, Amherst
Abstract
This paper describes a mostly automatic method for taking the out-
put of a single panning video camera and creating a panoramic
video texture (PVT): a video that has been stitched into a single,
wide field of view and that appears to play continuously and indef-
initely. The key problem in creating a PVT is that although only a
portion of the scene has been imaged at any given time, the out-
put must simultaneously portray motion throughout the scene. Like
previous work in video textures, our method employs min-cut opti-
mization to select fragments of video that can be stitched together
both spatially and temporally. However, it differs from earlier work
in that the optimization must take place over a much larger set of
data. Thus, to create PVTs, we introduce a dynamic programming
step, followed by a novel hierarchical min-cut optimization algo-
rithm. We also use gradient-domain compositing to further smooth
|
