Open Xuzzo opened 3 weeks ago
@Xuzzo Again, thanks for the contribution. Very nice that you spotted and fixed the height/width
mistake.
Regarding this feature, do you have a good example of it working? I only tried it quickly with the schnell model for 4 steps, and did not get any special result from it (I basically got back the input image with minor variations). Maybe it might not work as well with the distilled schnell
model??
Using the schnell
model I found it is quite important to tune the strength
parameter. For example:
Original image:
CLI:
python main_img2img.py --prompt "Luxury food photograph" --steps 10 --seed 2 --base-image image.png --strength 0.3
Result:
I found the dev to be a bit less sensitive to strength
.
Using the
schnell
model I found it is quite important to tune thestrength
parameter. For example:Original image:
CLI:
python main_img2img.py --prompt "Luxury food photograph" --steps 10 --seed 2 --base-image image.png --strength 0.3
Result:
I found the dev to be a bit less sensitive to
strength
.
Oh great, this is a very nice example! I think we should add this to the readme.
Btw, I have looked at your PR, and will share some comments later today. Thanks for the patience :)
Hi, this PR: