modelscope / modelscope/DiffSynth-Studio

Run the wan14B text-to-image model to generate a 21-frame video, with the first few frames being blurry and the later frames being normal.

Open
#519 2 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
13.1k
Forks
1.3k
Avg merge
13h 12m
Merged PRs (30d)
45

Description

Modify line 272 of wan_video.py to 21, then run wan_14b_text_to_video.py and observe that the first few frames are very blurry, while the later frames are normal.
Image

Image
first frame

Image
later frames

Image

first frame

Image
later frames
Are there any solutions or parameters that can be adjusted?

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

Reproduce the report by changing line 272 of wan_video.py to 21 and running wan_14b_text_to_video.py. Compare the first few blurry frames with the later normal frames, then identify whether an adjustable parameter or another cause explains the behavior; done means a supported solution or parameter adjustment is documented.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
machine-learning
Issue type
Bug
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.