Add gradients to intrinsics #801
Open
+763
−77
Conversation
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 join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.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.
This PR modifies FullyFusedProjection and FullyFusedProjection2DGS such that the intrinsics (
Ks) receive gradients.Am currently working on adding it to PackedProjection as well
Inspiration and derivation of gradients to K for 3DGS came from this paper.
2DGS gradients were derived by hand.
A simple test is added for now. It confirms that an "incorrect" K converges to the correct K from RGB gradients alone.
The results will look something like this:
However, while I am fairly confident, I am not 100% sure on the math of the gradients. Would be great if someone can double-check!
Not too sure on the design choice of overloading the
persp_proj_vjpfunction for 3DGS to support with/without Ks grads. Maybe we can change that.