PointNetGPD is an end-to-end grasp evaluation model to address the challenging problem of localizing robot grasp configurations directly from the point cloud.
First, let me thank you for your work and for making the code available!
I've been trying to run the generate-dataset-canny.py script with grasp_sample_method as 'gpg'.
However, in grasp_sampler.py, the sample_grasps function for GPG doesn't return a list of ParallelJawPtGrasp3D objects, as expected by the generate_grasps function.
I've tried to convert the sampled grasps to ParallelJawPtGrasp3D objects using ParallelJawPtGrasp3D.grasp_from_contact_and_axis_on_grid, but it always leads to a "Contacts not found" warning (using the YCB dataset).
Do you have any suggestions on how to get the GPG sampling to work?
Thank you in advance!
Hi!
First, let me thank you for your work and for making the code available!
I've been trying to run the generate-dataset-canny.py script with
grasp_sample_method
as 'gpg'. However, in grasp_sampler.py, thesample_grasps
function for GPG doesn't return a list ofParallelJawPtGrasp3D
objects, as expected by thegenerate_grasps
function.I've tried to convert the sampled grasps to
ParallelJawPtGrasp3D
objects usingParallelJawPtGrasp3D.grasp_from_contact_and_axis_on_grid
, but it always leads to a "Contacts not found" warning (using the YCB dataset).Do you have any suggestions on how to get the GPG sampling to work? Thank you in advance!