Shimingyi / MotioNet

A deep neural network that directly reconstructs the motion of a 3D human skeleton from monocular video [ToG 2020]
https://rubbly.cn/publications/motioNet/
BSD 2-Clause "Simplified" License
554 stars 82 forks source link

Global root position #55

Open asw91666 opened 1 year ago

asw91666 commented 1 year ago

Thank you very much for releasing your code. I had a few questions after reading your paper.

1) What exactly does global mean in the 'global root position' you mention in your paper? Do you mean the coordinates of the human pelvis in the camera coordinate system?

2) Did you use GT focal length to calculate the global root position as shown in Fig 10 of your paper? Or is there a way to calculate the global root position without knowing the focal length?