Why don't you use the normal/depth map renders?
Open
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.7k
- Forks
- 1.3k
- PR merge metrics
- No merged PRs in 30d
Description
Hi, why don't you use the normal/depth map renders but re-render those using .ply during training? Also, I see you didn't enable normal map loss which applies to predicted vertex normals, then how did you provide supervision to those attributes given that your pretrained checkpoints actually predict those attributes correctly?
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
No files, tests, or entry points are named. Trace the training and rendering paths for normal/depth maps, then inspect how pretrained checkpoints supervise predicted vertex normals; done means documenting a clear answer to both questions.
Written by the indexing model from the issue text.
Assessment
- Domain
- computer-graphics, machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100