OpenImagingLab / OpenImagingLab/FlashVSR
Training Details
Nobody has claimed this yet.
- Dominant language
- Python
- Stars
- 1.9k
- Forks
- 152
- PR merge metrics
- No merged PRs in 30d
Description
Thanks for your great work. I have a question: in Stage 1, there is no high-quality video used during training, so the model does not have the ability to repair videos. However, in Stage 3, the DMD loss is used to distill the model into a one-step model, where the training target is the latent features, and the decoder is not trainable. The gradient from the LPIPS loss will also affect the latents, which is inconsistent with the standard DMD setting.
What confuses me is whether high-quality videos are used in Stage 1 and Stage 2.
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
No file, test, or entry point is named. Start by tracing the Stage 1, Stage 2, and Stage 3 training configuration and the DMD/LPIPS loss paths; done means documenting whether high-quality videos are used in each stage and resolving the stated training-target question.
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
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 35/100