You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, I am training GauHuman on your dataset. I tried to reuse the Dataloader for ZJU, however the training result is always black. It seems that your dataset is using a different format with the ZJU Monocap dataset.
I only modify the data of camera translation by multiplying with camera_scale.pkl. The way I convert original data of MVHuman to ZJU format is:
new_params: taken from smpl/smplx.
new_vertices: taken from smpl_param
cams: taken from camera_intrinsics, camera_extrinsics
kpts2d: taken from openpose
Questions/notices:
The SMPL parameters has shape of (1, 87), while it is (1, 72) in ZJU.
Is the camera convention in MVHuman the same as in ZJU?
How can I convert the MVHuman data to the exact format of ZJU?
Thank you!
The text was updated successfully, but these errors were encountered:
Hi, thank you for your work.
Currently, I am training GauHuman on your dataset. I tried to reuse the Dataloader for ZJU, however the training result is always black. It seems that your dataset is using a different format with the ZJU Monocap dataset.
I only modify the data of camera translation by multiplying with
camera_scale.pkl
. The way I convert original data of MVHuman to ZJU format is:new_params
: taken fromsmpl/smplx
.new_vertices
: taken fromsmpl_param
cams
: taken fromcamera_intrinsics
,camera_extrinsics
kpts2d
: taken fromopenpose
Questions/notices:
Thank you!
The text was updated successfully, but these errors were encountered: