Other existing methods of automatic stereo video conversion
from single view video typically work by reconstructing a dense
depth map using parallax between frames [5], or structure from motion
[6]. However, these methods require static scenes and specific
camera paths, and in cases where parallax does not exist in a video
sequence, such as with a rotating camera, these methods would not
work. Our method produces high quality automatic stereo conversion
without assuming static content.
2011 18th IEEE International Conference on Image Processing
978-1-