Human pose prior in Volumetric triangulation

[Oruka94]

Hi.
You mentioned several times in the paper that your volumetric triangulation method could learn a human pose prior. But I can not see that thing clearly. Could you please tell more detail about it? At what step is it done?

[urs-waldmann]

I am also interested in that topic. Thank you very much in advance for your help, time and clarification.