how does the to3d func work?

[longbowzhang]

Hi, thanks for sharing the great work.
I have a question about heatmap. I noticed that you encode the distance of the joints to the camera cam_dist = np.sqrt(joint.x3d ** 2 + joint.y3d ** 2 + joint.z3d ** 2) , why not simply encode the z-depth value of the joints joint.z3d? any particular reason?

Besides, could you elaborate a little bit on

LoCO/utils.py

Line 97 in 7ebaca7

def to3d(x2d, y2d, cam_dist, fx, fy, cx, cy):

, basically what does this k value mean?

LoCO/utils.py

Line 113 in 8371ef0

k = (-1) * np.sqrt(

[fabbrimatteo]

Hi @longbowzhang,

Thank you for your interest in our work.

You are right, it is much simpler to encode the z-depth value directly. In fact we used this approach for Panoptic and Human3.6 as described in the paper's supplementary material.

As for the to3d function, it has been obtained by solving the following system w.r.t. the 3D coordinates:

k is just a recurrent coefficient calculated one time and used to obtain each 3D coordinate.