microsoft / microsoft/MoGe

normal head supervision

Open
#149 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
JavaScript
Stars
3k
Forks
234
PR merge metrics
No merged PRs in 30d

Description

Hi,

Thanks for great work! I saw on normal.md there is "For convenience, we did not collect ground-truth normal maps for training. Instead, we derived surface normals from the depth map and camera intrinsics." May I have you tried direct gt normal map supervision for normal head? which design choice will give you better results?

Thanks!

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

Start by reading normal.md, especially the section describing normals derived from the depth map and camera intrinsics. Investigate the normal head's training supervision and compare that design with direct ground-truth normal-map supervision; done means documenting which choice performs better and why.

Written by the indexing model from the issue text.

Assessment

Domain
computer-vision, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.