Hi, beginner here trying to understand the denoiser workflow.
I've gotten so far that I understand that we have denoising as a post process effect via the Arnold Imager not suited for animations. Then we have the Arnold Denoiser that works on rendered .exr images with a specific AOV, and this do work with animation.
I tried running the Denoiser several times without any result, the new image sequence was exactly the same as the input. Then I tried changing the default filter to gaussian from box and all of a sudden it worked. Why is this? What is happening here? What am I not understanding in how gaussian calculates the sub-pixels that makes it required for denoising?
Anyone care to explain this process?
The beauty has to use the same filter as the other denoiser AOVs (N, Z, variance, denoise_albedo). Those AOVs are created by C4DtoA for you, and they use the gaussian filter.
Thanks Stephen.
This info should be added to the documentation, hard to figure this out through the help docs.
Can't find what you're looking for? Ask the community or share your knowledge.