search

nv-nguyen / cnos

[ICCV 2023 R6D] PyTorch implementation of CNOS: A Strong Baseline for CAD-based Novel Object Segmentation based on Segmenting Anything and DINOv2
MIT License
204 stars 22 forks source link

Pixel to point correspondances with renderer #14

Closed fabiopoiesi closed 1 year ago

fabiopoiesi commented 1 year ago

Hi, can the renderer provide correspondences between pixels on a rendered image and the points on the CAD model?

nv-nguyen commented 1 year ago

The renderer does not provide it directly but you can create a function to have from the depth map and object poses (first flip the depth into 3D point cloud (in camera coordinate) then transform them into the object coordinate)