Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Fix point cloud jitter and quantization issues for additive refinement pnts #5324

Merged
merged 14 commits into from
Apr 9, 2023

Conversation

DStradley
Copy link
Contributor

@DStradley DStradley commented Mar 31, 2023

Fixes #5333

Investigating problems with a user point cloud showed jitter and gaps in the displayed data, both indicating quantization issues. The point cloud was PNTS format, but using additive refinement. Also, the first 13 pnts files were unquantized, but all the rest were quantized.

Found that we were quantizing the unquantized tiles, so first changed that to generate and use unquantized point cloud graphics (for PNTS), but that by itself did not solve the problem. Investigating further, I found that the RTC_CENTER in these tiles was large enough to cause quantization issues the way it was added into the vertex shader points. I removed that offset from the vertex shader points, then added it into a transform for the tile instead, and that resolved the problem.

Had to adjust batch range by RTC_CENTER since it is in the transform now (for locate, etc.). Also had to create a range for unquantized points for the voxelSize calculation, since the qParam range (scale of 1) would not work for that.

TODO

  • Run ImageTests

@DStradley
Copy link
Contributor Author

ImageTests ran and the only differences were considered inconsequential ones from now supporting unquantized pnts data.

@DStradley DStradley closed this Apr 7, 2023
@DStradley DStradley reopened this Apr 7, 2023
@DStradley DStradley marked this pull request as ready for review April 7, 2023 20:21
@DStradley DStradley requested review from kabentley, bbastings and a team as code owners April 7, 2023 20:21
@DStradley DStradley requested a review from pmconne April 7, 2023 20:22
Copy link
Member

@pmconne pmconne left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Run rush change and add the files created in common/changes.

@pmconne
Copy link
Member

pmconne commented Apr 7, 2023

I am still getting the extract-api error

The error is telling you what you need to do to fix it.

@DStradley
Copy link
Contributor Author

I did "npm run extract-api" in the core\extension folder (step 1), and it generated no modified files anywhere, so can't do step 2.

@DStradley DStradley enabled auto-merge (squash) April 7, 2023 22:06
@DStradley
Copy link
Contributor Author

Don't know what to do about the repeatable MacOS security error.

@pmconne
Copy link
Member

pmconne commented Apr 9, 2023

Don't know what to do about the repeatable MacOS security error.

As yet there is no fix available. Nothing to do with your changes.

@pmconne pmconne disabled auto-merge April 9, 2023 09:22
@pmconne pmconne merged commit 7de7e53 into master Apr 9, 2023
@pmconne pmconne deleted the djs/fix-point-jitter branch April 9, 2023 09:22
@DStradley
Copy link
Contributor Author

@Mergifyio backport release/3.7.x

@mergify
Copy link
Contributor

mergify bot commented Apr 19, 2023

backport release/3.7.x

✅ Backports have been created

mergify bot pushed a commit that referenced this pull request Apr 19, 2023
…t pnts (#5324)

Co-authored-by: Paul Connelly <[email protected]>
(cherry picked from commit 7de7e53)
aruniverse added a commit that referenced this pull request Apr 19, 2023
…t pnts (backport #5324) [release/3.7.x] (#5421)

Co-authored-by: DStradley <[email protected]>
Co-authored-by: Arun George <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Fix point cloud jitter caused by quantization
2 participants