search

jfzhang95 / PoseAug

[CVPR 2021] PoseAug: A Differentiable Pose Augmentation Framework for 3D Human Pose Estimation, (Oral, Best Paper Award Finalist)
MIT License
368 stars 57 forks source link

Regarding the 2D input #20

Closed jacksoncsy closed 3 years ago

jacksoncsy commented 3 years ago

Hello,

Sorry to bother you again, may I know which joint is not needed for your 2D input? Is it head, or nose? I understand that your input and output are both 16 joints, and nose is not predicted by the model.

Many thanks, Shiyang

Garfield-kh commented 3 years ago

Hi, Shiyang,

Welcome~ Both 2D pose and 3D pose follow the same joint setting so that the virtual camera projection (3D pose 16x3 -> 2D pose 16x2) can be used as augmentation data directly.

jacksoncsy commented 3 years ago

Hey Kehong,

So the nose is excluded in both setting then? Thanks so much!

Cheers, Shiyang

Garfield-kh commented 3 years ago

Hi Shiyang,

Yes. :)