thodan / obj_pose_eval Goto Github PK

When I run "eval_rotating_cup.py" sample code
it shows this error:
ModuleNotFoundError: No module named 'obj_pose_eval'

I am trying to render a mesh to color and depth image using the given code on a textured mesh, while the code didn't throw an exception, the generated color and depth image are all blank.

At first, I thought it is a translation error that the mesh is not within the FOV of the camera, but I make sure to center the mesh and adjust the translation but still nothing. I would appreciate if you can give me some suggestions, I have included the link to the mesh I am using.

https://drive.google.com/open?id=1mXj6XBCLGHdjKUuBGUWR6ew2BlUGKGka

Thanks!

Hello,
Is it possible to test the Nvidia deep pose models same as the eval_rotating_cup?

Thank you!

Hey, thanks for creating this tool, could you please also add python3 support?