Depth eval
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 397
- Forks
- 34
- PR merge metrics
- No merged PRs in 30d
Description
Hi DAVID team,
Congrats on the great work!
We're evaluating our method on Goliath too, to get of feeling of how our method perform in the fair comparison setting. Currently our performance is similar but also quite varied to yours, e.g. we have very similar AbsRel, but very different RMSE (0.1 times lower).
The difference looks like a scale problem. So I'm wondering, could you share more details or code on your depth evaluation? For example, how did you render the ground-truth depth from Goliath, what kind of alignment algorithm are you applying?
In our case, we align the predictions to the ground-truth, and then perform evaluation. I think your paper is doing the same? (according to your descriptions in the github)
Any guidance will be appreciated! thanks!
Best,
Letian
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
Start by locating the repository's depth-evaluation workflow and the handling of Goliath ground-truth depth, since the issue names no files or tests. Document or share the rendering and prediction-alignment procedure, including how the reported metrics are produced, so another evaluator can reproduce the comparison.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-vision, machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100