The missing encoder in the structure generation
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 13.7k
- Forks
- 1.3k
- PR merge metrics
- No merged PRs in 30d
Description
The paper said in the sparse structure generation process, we need to use a VAE to compress the occupancy grid to reduce memory usage in attention computation, but the code below only shows the decoder, but no encoder. This is confusing, and I may have overlooked details in the code or paper. Can someone help me with this problem?
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 with the cited section of trellis/pipelines/trellis_image_to_3d.py at lines 177-191, then compare its sparse structure generation flow with the paper's VAE description. Trace the referenced pipeline components to determine whether occupancy-grid encoding happens elsewhere or is missing. Done means documenting the encoder's location and role, or clearly recording the confirmed discrepancy.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning
- Issue type
- Documentation
- Difficulty
- 4/5
- Estimated time
- 3-5 days
- Activity status
- Stale
- Clarity
- Needs clarification
- Newbie friendliness
- 25/100