theotherphil

I've pushed some changes to https://github.com/image-rs/imageproc/tree/seams to reallocate slightly less. How does this compare to the existing implementation in your tests?

Excellent! (To both sets of results)

Ah, great. I’ll have another go at speeding my version up a bit more over the weekend and I’ll use your benchmarking harness.

I’m still interested in following up on this, but “this weekend” was a bit optimistic.

Thanks for the PR. This is definitely a useful function to add. The image you're committing for the regression test appears to have a black bar running down the left...

criterion-plot uses an interesting hack - the doc generation first runs cargo test and then links to the target directory from the docs. This lets you include images in the...

I'm actually not sure how to add pictures to the documentation in a way that will work when rendered on docs.rs. Adding some images to the README is easier -...

Sorry for the slow replies. I didn't have any particular plans for what to show - I was imagining just picking a few functions that produce interesting looking results and...

Thanks for the link. (Turns out I didn't get round to looking at this after all. I'm still interested in implementing this functionality, but I'm not sure when it'll get...

Thanks, I'll take a look. I know [VLFeat](http://www.vlfeat.org/index.html) allows zero or continuity padding. I've never looked into how this sort of thing is done in graphics libraries before.