| 摘要 |
A technique for compactly representing the surface radiance characteristics of an object based on acquired images of the object includes calculating surface radiance functions, which are then partitioned, based on a mesh topology, into vertex radiance functions. The vertex radiance functions are then reparameterized, on a vertex-by-vertex basis, to local coordinates systems. The reparameterized vertex radiance functions, which may be six-dimensional, are then decomposed into a plurality of lower dimensional functions that may be compactly stored.
|
