1. A method for encoding a video sequence with a non-static viewpoint, said video sequence having a plurality of frames and
comprising a background composite constructed from the plurality of frames and foreground regions for individual frames of
the plurality of frames, comprising the step of:
encoding said video sequence based on balancing bits per pixel for said background composite with bits per pixel for said
foreground regions to achieve similar quality between the background composite and the foreground regions in a reconstructed
video sequence, wherein said bits per pixel for said background composite are based on a number of bits for texture encoding
of said background composite and a number of bits for warp points of said background composite, and said bits per pixel for
said foreground regions are based on a number of bits for texture encoding of said foreground regions and a number of bits
for shape encoding for said foreground regions.
|