The data of humanml is of 153 dimension rather than 322 dimension.

[weihaosky]

Hi, thanks for this excellent work!

When I try to transfer the data to tomato representation with python raw_pose_processing.py, I find the data in face_motion_data/smplx_322/humanml or GRAB or EgoBody is of 153 dimension rather than 322 dimension, such that it cannot be processed by python raw_pose_processing.py.

How should I get the data of 322 dimension for dataset humanml, GRAB, and EgoBody?

Many thanks!

[linjing7]

Hi, have you preprocess the mocap data following this instruction?

[weihaosky]

Hi, have you preprocess the mocap data following this instruction?

No, I download the motionx_face_motion_data.zip, which contains face_motion_data/smplx_322/humanml&GRAB&EgoBody
I did not do the preprocess from AMASS dataset.