Known camera poses

[ngailapdi]

Hi,

Thank you for the great work. I know that Human3R is based on CUT3R, and CUT3R does not support known camera parameters. I wonder if you have any suggestions for integrating known camera poses into Human3R pipeline?

[fanegg]

Hi @ngailapdi, thank you for your interest! You can try global alignment with camera ground truth:)