microsoft / microsoft/TRELLIS

Forward pass of rectified flow model

Open
#164 1 comment 3 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
13.7k
Forks
1.3k
PR merge metrics
No merged PRs in 30d

Description

I'm reimplementing the forward process of your rectified flow model but I am not able to find details in the paper and the implementation.

Could you elaborate more regarding converting a mesh into the active voxel representation and how you generate the noise for the rectified flow model during sparse structure generation and later during structured latents generation? Also, could you provide hyper-parameters that you use during the process?

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

No files, tests, or entry points are named. Read the paper and implementation sections covering mesh-to-active-voxel conversion and rectified-flow noise for sparse structure and structured latent generation. The documentation should explain both forward processes and list the hyperparameters used.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
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

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.