search

una-dinosauria / 3d-pose-baseline

A simple baseline for 3d human pose estimation in tensorflow. Presented at ICCV 17.
MIT License
1.42k stars 353 forks source link

Understanding the Extrinsic Parameters of Human3.6M #183

Closed wenwu35 closed 4 years ago

wenwu35 commented 4 years ago

Hi, I am trying to make sense of the extrinsic parameters of the Human3.6M dataset. I first thought the extrinsics (R, t) of each of the 4 cameras pointed to one single world coordinate center, but it didn't seem to be right after I reconstructed the cameras views using Human3.6M's extrinsic parameters (image below) - the cameras were not placed at the 4 corners of the capture space as described by Lab Setup.

Just wondering if anyone knows how each camera of Human3.6M set their O points, and are they the same point in the real physical world? Many thanks!

image

una-dinosauria commented 4 years ago

That sounds like an excellent question for the authors of the Human 3.6M dataset.