Congrats on the amazing paper! The performance of the method looks awesome.
May I ask what FLAME-tracking method did you use to obtain the camera pose/flame parameters etc? In the paper it says you used the same algorithm for all datasets. Is it the preprocessing pipeline developed by INSTA/PointAvatar, or is it something you developed on your own?
Hi,
Congrats on the amazing paper! The performance of the method looks awesome.
May I ask what FLAME-tracking method did you use to obtain the camera pose/flame parameters etc? In the paper it says you used the same algorithm for all datasets. Is it the preprocessing pipeline developed by INSTA/PointAvatar, or is it something you developed on your own?
Many thanks