¡@
c++ Programming Glossary: convolutionFindChessboardCorners cannot detect chessboard on very large images by long focal length lens http://stackoverflow.com/questions/15018620/findchessboardcorners-cannot-detect-chessboard-on-very-large-images-by-long-foca to find the corners have fixed size and that size of convolution mask may be too small to detect your corners the image may actually..
How do I gaussian blur an image without using any in-built gaussian functions? http://stackoverflow.com/questions/1696113/how-do-i-gaussian-blur-an-image-without-using-any-in-built-gaussian-functions easy. It is done in exactly the same way as any other convolution filter. The only difference between a box and a gaussian filter..
Fast bignum square computation http://stackoverflow.com/questions/18465326/fast-bignum-square-computation xx k 0 NTT fourier_NTT ntt ntt.NTT yy xx n init NTT for n convolution for i 0 i n i yy i modmul yy i yy i ntt.p INTT ntt.INTT xx yy..
modular arithmetics and NTT (finite field DFT) optimizations http://stackoverflow.com/questions/18577076/modular-arithmetics-and-ntt-finite-field-dft-optimizations y x N NTT y N no recompute of constants use last N modular convolution y z .x for i 0 i n i y i ntt.modmul z i x i ntt.INTT x y x N.. x y x N INTT y N no recompute of constants use last N x convolution of original x .y some measurements before optimizations non..
Optimized float Blur variations http://stackoverflow.com/questions/7860575/optimized-float-blur-variations S O X is not separable. You need to actually compute a 2D convolution for these. As for performance you can speed up your 1D averages.. separable filters your best bet performance wise is fast convolution which is FFT based. Checkout www.dspguide.com. If I recall correctly.. there is even a chapter on how to properly do fast convolution using the FFT algorithm. Although they explain it for 1 dimensional..
|