I created this to fix my some pain points when making the character walk:
Sometimes the character walks but moves backward.
Sometimes the character walks but moves forward for a few step then move backward,
Sometimes the character walks in place (walk but not go anywhere)
Sometimes the steps go out of sync, making the movement look unnatural.
.
V3.0
Main Fix: Resolves the moonwalk issue. Based on my experiments, it’s a lot better now.
Improvement: Fixes cases where some images caused the model to stubbornly refuse to walk.
Strength: 0.7-1.0
Trigger Word (Important):
wlkng,
The subject walks casually alongside the viewer, with the camera following the subject’s movement.
(Thanks to @Corbe for proposing the prompt idea that helped fix the moonwalk issue.)
Description
V2.0
Train with a doubled dataset to
fix cases where some images make the model stubbornly refuse to walk.
Strength: 0.7 - 1.0
Trigger word (camera from front) :
wlkng,
she is walking towards camera,
Trigger word (camera from behind) :
(strength for behind should be 0.75)
wlkng,
she is walking,
Trigger word (camera from side) :
wlkng,
she is walking,
.
.
NOTE:
You might notice that sometimes the girl’s body gets pulled back to the original position after walking for a while.
This happens because of Wan2.2 itself, and also because most of the dataset clips are only 4–5 seconds long.
But even when I tried training with 25 clips of 12 seconds each, the results improved alot, but the pulling effect didn’t completely disappear.
Solutions to reduce the pull-back effect:
Avoid using fully front-shot images – the background tends to pull her backward.
Use side view or 45° front-side view images – I tried this and it worked well.
(Optional) Try a new seed or a different input image – sometimes this helps smooth out the walking motion.
(but) if you don't care about position movement you just care about girl walking motion and you okay if her walking in place, so you can try this "the girl is walking on treadmill"