github/ControlNet
1
1
Fork 0
mirror of https://github.com/lllyasviel/ControlNet.git synced 2026-04-24 03:00:54 -04:00
Code Issues Packages Projects Releases Wiki Activity

guess mode

Browse Source
This commit is contained in:
lvmin
2023-02-20 14:04:00 -08:00
parent 68616f1cd9
commit 4e151d0ba0
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
README.md
View File

@@ -258,6 +258,7 @@ Without prompt, the HED seems good at generating images look like paintings when
As you can see below, because **non-prompted mode is extremely difficult**, even the very deep ControlNet may also generate failure cases. (In the example below, the right-bottom one seems ok while others look meh.) Because of this, the non-prompted mode seems a nice metric to test the capability of different encoder-based methods to control stable diffusion.
**No prompts. No "positive" prompts. No "negative" prompts.**
![p](github_page/uc7.png)
Note that in the guess mode, you will still be able to input prompts. The only difference is that the model will try harder to guess what is in the control map even if you do not provide the prompt. Just try it yourself!