Re: Projective image transformations
Owen Taylor <[email protected]> Thu, 12 Sep 2002 20:15:42 -0400 (EDT)
| Newsgroups | gmane.comp.xfree86.render |
|---|---|
| Message-ID | <[email protected]> |
Keith Packard <[email protected]> writes: > > Edge conditions are definitely the hard part of the specification; consider > > a straight scale of a solid rectangle from 100x100 pixels by a ratio > > of 75/100. The desired result is obvious ... a 75x75 pixel solid square > > with hard edges. > > Both nearest neighbor and bilinear interpolation will yield this result. This example and my last mail were making things too complicated. Lets take a *very* simple example - scaling a 1x1 solid colored constant square to 3x3 with bilinear interpolation and compositing it on a 3x3 destination image. Originally we have alpha values: 1 When scaled to 3x3 with bilinear interpolation, using the "all pixels outside the image are transparent rule", we get: 0.44 0.66 0.44 0.66 1 0.66 0.44 0.66 0.44 I don't think this is a terribly useful result. (You don't have to get as complicated as my partial-convolution suggestion to fix this ... a simpler rule is to simply clamp source pixel coordinates to the bounds of the original image.) Regards, Owen