search

RobotLocomotion / pytorch-dense-correspondence

Code for "Dense Object Nets: Learning Dense Visual Object Descriptors By and For Robotic Manipulation"
https://arxiv.org/pdf/1806.08756.pdf
Other
557 stars 133 forks source link

Find corresponding pixel in the image of vertices in fusion_mesh.ply #214

Closed lix4 closed 4 years ago

lix4 commented 4 years ago

Hello, maybe this is irrelevant with this project. I am trying to find x and y coordinate the image of a vertex in mesh. I just use Ktransform_matrix(X,Y,Z,1) (vertex coordinate in mesh.). But I got incorrect pixels. Is there anything wrong with doing this? Thank you!

manuelli commented 4 years ago

Projecting a world point into camera frame is just done using the pinhole camera model, see OpenCV documentation. This is a simple transformation to write and the purpose of this repo is not to be a library for projective geometry.