MotrixLab / MotrixLab/ADHMR

About performance

Open
#6 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
47
Forks
0
PR merge metrics
No merged PRs in 30d

Description

Hello, I used the checkpoint from scorenet and OneDrive for ADHMR. Then I get the following results, which are much lower than the results reported in the paper.

3dpw dataset

MPVPE (Min): 62.61 mm
MPVPE (Select):79.96 mm
MPVPE (Zero Noise): 80.22 mm
MPVPE (Hypoes Mean):80.27 mmMPJPE (Min): 54.19 mm
MPJPE (Select): 68.87 mm
MPJPE (Zero Noise): 68.46 mm
MPJPE (Hypoes Mean): 68.59 mm

PA MPJPE (Min):31.97 mm
PA MPJPE (Select): 42.04 mm
PA MPJPE (Zero Noise): 41.93 mm
PA MPJPE (Hypoes Mean):41.99 mm

I'm wondering if I missed some steps?

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 reproducing the reported 3DPW evaluation with the scorenet and OneDrive checkpoints linked in the issue, then compare the MPVPE, MPJPE, and PA MPJPE results with the paper. The issue does not name evaluation files, tests, or a specific missing step; done would mean identifying and documenting the cause of the metric discrepancy.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
computer-vision, machine-learning
Issue type
Bug
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.