About coordinate transform from camera frustum space to 3d world

chenyilun95 / DSGN2

DSGN++: Exploiting Visual-Spatial Relation for Stereo-based 3D Detectors (T-PAMI 2022)

Apache License 2.0

79 stars 8 forks source link

About coordinate transform from camera frustum space to 3d world #2

Closed QitaoZhao closed 2 years ago

QitaoZhao commented 2 years ago

Could you please explain what cu and cv mean in the formula? Besides, I'm curious about the derivation of this formula. I wonder if you could provide some references. Thanks for your nice work.

chenyilun95 commented 2 years ago

They are the principal points of camera reference. More details are referred to camera projection geometry.