This repository was archived by the owner on Jun 2, 2025. It is now read-only.
Use GPU for otherwise very slow "Image Film Grain" node (if available)#462
Open
ClownsharkBatwing wants to merge 3 commits intoWASasquatch:mainfrom
Open
Use GPU for otherwise very slow "Image Film Grain" node (if available)#462ClownsharkBatwing wants to merge 3 commits intoWASasquatch:mainfrom
ClownsharkBatwing wants to merge 3 commits intoWASasquatch:mainfrom
Conversation
Owner
|
Oh hey thanks. Yeah a lot of the noise/image stuff could be moved to torch. Will most. Some stuff I found PyTorch implementations of kinda inferior. Like doing sharpening. |
Owner
|
Did you mean to include a "CLIPSeg2" in here? |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The "Image Film Grain" node was effectively generating large noised images one pixel at a time looping through randint() calls, which is a very slow process. This forked version uses pytorch and is much faster - on a 5950x + 4090 system, 1024x1024 images went from taking at least 30 seconds to maybe 0.2 seconds. The benefits are even greater with larger images, which drop from taking several minutes to under a second.