Feature Request: Faster DCT #245
Description
Is your feature request related to a problem? Please describe the problem.
DCT is slow for big sizes because it is a brute force implementation and the processing complexity is O(n^2) - initialisation is either O(n^2) or O(n^3)....
Describe the solution you'd like to see.
A faster DCT (possibly using FFT) and possibly implemented by @lewardo
Describe alternatives you've considered
Getting a supercomputer
Additional context
PA agrees with me