Mirrors/AnimateDiff
1
0
Fork 0
mirror of https://github.com/guoyww/AnimateDiff.git synced 2025-12-16 16:38:01 +01:00
Code Issues Packages Projects Releases Wiki Activity

Update README.md

Browse Source
This commit is contained in:
Ceyuan Yang
2023-11-13 14:50:06 +08:00
committed by GitHub
parent 397ec30d15
commit 8d191074a1
1 changed files with 3 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
README.md
View File

@@ -3,7 +3,8 @@
Now you can generate high-resolution videos on SDXL **with/without** personalized models. Checkpoint with better quality would be available soon. Stay tuned.
## Somethings Important
- Generate videos with high-resolution (we provide recommended ones) as SDXL usually leads to worse quality for low-resolution images.
- Generate videos with high-resolution (we provide recommended ones) as SDXL usually leads to worse quality for low-resolution images.
- Quality strongly depends on the sampler and #sampling stepes (we use EulerDiscrete sampler for 100 steps). Be careful.
- Follow and slightly adjust the hyperparameters (e.g., #sampling steps, #guidance scale) of various personalized SDXL since these models are carefully tuned to various extent.
## Model Zoo
@@ -148,4 +149,4 @@ n_prompt:
Then run the following commands.
```
python -m scripts.animate --exp_config [path to the personalized config] --L [video frames] --H [Height of the videos] --W [Width of the videos] --xformers
```
```