Repository navigation
Conversation
The density and alpha compositing helpers are annotated for arbitrary leading ray batch dimensions, but create identity prefixes from only the first batch axis. The alpha helper also accumulates and slices along dimension 1. Unbatched rays and image-shaped ray batches consequently fail or composite along the wrong axis.
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
Volume-compositing weight helpers must preserve all leading ray-batch dimensions rather than relying on a flat batch. Incorrect singleton construction and concatenation dimensions can break multi-axis batches.
Change
Build leading singleton and transmittance tensors with the full ray-batch shape, preserving the sample axis and dtype across both alpha- and density-based weight paths.
Regression coverage
New CPU regressions exercise scalar, one-dimensional and multi-dimensional ray batches in float32/float64, comparing alpha-based and density-based transmittance against direct expressions. Historical baseline-versus-fix focused results were prepared on the source branch; they were not rerun in this handoff.
The upstream contribution guide requires the full
ns-dev-testchecks to pass. Existing Afloat16 Nerfstudio PRs show Core and Docker failures; this new PR has no own hosted CI. No GPU/rendering-quality benchmark is claimed.