The latent VAE failed to reconstruct the mesh
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.7k
- Forks
- 1.3k
- PR merge metrics
- No merged PRs in 30d
Description
Dear Sir,
I test the latent VAE but find that only gs decoder can reconstruct the input, while mesh decoder and rf decoder can not reconstruct the input mesh. To demonstrate, the reconstructed mesh is rendered as:
GS:
https://github.com/user-attachments/assets/b17a74d4-a5b2-4112-8129-5570b0253ab1
https://github.com/user-attachments/assets/6d7d248b-2745-43cc-b686-c861556e51a9
https://github.com/user-attachments/assets/c07c8420-a8e2-4e1d-aeee-1253c25ce921
MESH:
https://github.com/user-attachments/assets/847ed68d-5955-4806-9753-5ae07af663c1
https://github.com/user-attachments/assets/d8a17b50-aa7e-464c-ae9e-c093d5972ad7
https://github.com/user-attachments/assets/0a90ab03-296d-4b0e-a9b1-3143f83d354b
RF:
https://github.com/user-attachments/assets/087dcc88-ce3b-4831-95cb-0db4a3d86c3c
https://github.com/user-attachments/assets/7a92d265-cddb-40d8-9bab-cf314e6688ae
https://github.com/user-attachments/assets/685c0eb7-2ef1-4eae-a2e2-bdbd8eb1b636
From these videos, we can observe that only the GS decoder can reconstruct the input, while the MESH and RF failed.
The code I used is very simple:
I wonder why this happened. Could you help me solve this problem?
Thanks.
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 reproducing the latent VAE run using the code shown in the issue and compare the GS, MESH, and RF decoder outputs. The issue is resolved when the MESH and RF decoders reconstruct the input mesh correctly, with the underlying cause identified and documented.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- computer-graphics, machine-learning
- Issue type
- Bug
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100