Morphable 3D-Mosaics: A Hybrid Framework for Photorealistic Walkthroughs of Large Natural Environments
Figure 7
We show rendered views of a morphable 3D-model during
the transition from one key-position (corresponding to image in Figure 4) to another key-position
(corresponding to image in Figure 4). On t(a) we show the result
produced when no geometric morphing is applied to points in (notice the resulting discontinuity, i.e.,
hole). On (b) we render the same view, but this time we also apply
geometric morphing to points in (no discontinuities/holes exist in this case).
A zoomed version of this view is also shown on (c). Although, as
mentioned, there is no geometric discontinuity, yet there is a difference in
texture resolution between the left part of the image (points in ) and the right part (points in ), since only points of the latter part are
morphed photometrically.