hi all,
i’m looking for information about antialiasing filter.
but on 2D image after ( or since ) a warp deformation.
i know the standart box filter, but i have to compute big image, so it’s so difficult to use it, i have lot of problem with RAM
do you have URL or info, on antialising methode, which use more process ( CPU ) but less memory than box methode ?
thx for your help
( sorry for my poor english )
Kermit
Where is your algorithm using so much memory? The filter algorithm should only have to use enough memory to hold the kernel pixels
Hey,
You just want to antialias a large 2D image? Well if the picture is too large to fit into memory and swap (that thing has to be huge!), I suppose you could just read say 3 horizontal lines of it at a time and then perform a simple antialias of the second line. (For example by adding together the values of each pixel’s neighbors and taking their average. It’s probably good if you make the average weighted as well so that the ‘center’ pixel has the biggest impact.)
Then just write the antialiased lines back to a new file as you go.
Maybe I misunderstood your question. 
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.