Open
Description
Hi thanks for the lib!
Context: I have a color image Array2<RGBA8>
, and want to visualize it using rust-numpy lib (a binding to python numpy). However, .to_pyarray()
complains, so I should convert it into Array3<u8>
.
Question: How can I do that - what is the correct/fast way to do so? IMHO the fastest way should even have no memory copy at all, since RGBA8 is compact 4-byte data. But one memory copy is also acceptable.
I have come up with a very slow method but I hope there are other ways: extract Array2<RGBA8>
into four Array<u8>
for each of 4 channels, and stack!
them together. Its performance seems to be quite slow.