CompVis / CompVis/EDGS

Question about SH coefficient initialization implementation

Open
#24 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Jupyter Notebook
Stars
742
Forks
59
PR merge metrics
No merged PRs in 30d

Description

Hi, thank you for releasing the code to this awesome work!

I was reading Sec. 3.5 of the paper, where it is stated that the initial spherical harmonics coefficients are estimated for each sampled Gaussian by collecting RGB observations from multiple views and solving a linear least-squares system.

However, while looking through the current repository, I could not find this SH initialization step. In source/corr_init.py, it seems that the initial color is assigned to the DC component using the reference image color, while the remaining SH coefficients are initialized to zero.

Image

Could you clarify whether the SH coefficient initialization described in Sec. 3.5 is currently implemented somewhere else in the repository, or whether it is omitted from the current public release? If it is omitted, is this planned to be included in a future code update?

Thank you again for your work and for making the code available!

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Read Sec. 3.5 of the paper and inspect source/corr_init.py, especially the initialization of the DC and remaining spherical harmonics coefficients. Search the repository for other SH initialization code and compare it with the described multi-view least-squares procedure. Done means determining whether the procedure exists elsewhere, is omitted from the release, or has a stated future-update plan.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
computer-graphics
Issue type
Documentation
Difficulty
4/5
Estimated time
3-5 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.