TMElyralab / MusePose

MusePose: a Pose-Driven Image-to-Video Framework for Virtual Human Generation

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Does not recognize more than one person

SWAPv1 opened this issue · comments

commented

I tried using a two person dance video and the dwpose only recognizes one person each frame.

Is there a way to modify the alignment to recognize multiple people?

Thanks.

That's because the model only considers one person during training. And I made the code only render the most obvious person detected. However,even if you render both of them, the model may not be able to drive two people to dance. The results may look strange.