How to prepare the 3D scan data?

[cocoshe]

You are using some kind of device or using something like PIFu/ICON? Any ways to get the 3D scan input data from monocular video(with only RGB)？

[lizhe00]

Our scans are collected using a dense multi-view DSLR rig. Models reconstructed from monocular videos are low-quality and incomplete, thus the learned avatar quality also degenerates.

[cocoshe]

Our scans are collected using a dense multi-view DSLR rig. Models reconstructed from monocular videos are low-quality and incomplete, thus the learned avatar quality also degenerates.

Thanks, I got it. Appreciate to your great work!