it can't work when somebody do Handstand in the pic

DenisTome / Lifting-from-the-Deep-release

Implementation of "Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image"

https://denistome.github.io/papers/lifting-from-the-deep

GNU General Public License v3.0

450 stars 132 forks source link

it can't work when somebody do Handstand in the pic #1

Closed wxy656 closed 7 years ago

wxy656 commented 7 years ago

If the person in the picture is not standing (such as push-ups or inverted), it can not be promoted to 3d. Will you follow up on this project so that it can support the actions of various scenes?

DenisTome commented 7 years ago

Hi, the probabilistic model that we provide for 3D lifting can be adapted to handle those kind of poses. The main priority right now is to provide a system working on the main data-sets for 2D and 3D human pose estimation. Some of these, like Leeds (http://www.comp.leeds.ac.uk/mat4saj/lsp.html), have poses where the person is standing upside-down, hence we'll have to update the "3D lifter".