Previously, I had used Nvidia Deep Learning Dataset Synthesizer (NDDS) for creating my dataset for 3D object pose estimation. I was able to scatter objects into a plane and also randomize the number of scattered objects between a chosen min and max. I searched through replicator documentation but couldn’t find anything that could show this functionality and how to do that.
Are you able to add this functionality to scatter_2d or if it already exist can you please guide me how to use it?
Here’s an example from offline_generation.py
from
(base) mona@ard-gpu-01:~/.local/share/ov/pkg/isaac_sim-2022.2.0$ ls standalone_examples/replicator/offline_generation.py -rw-r--r-- 1 mona mona 21K Feb 2 21:33 standalone_examples/replicator/offline_generation.py
def scatter_boxes():
cardboxes = rep.create.from_usd(
prefix_with_isaac_asset_server(CARDBOX_URL), semantics=[("class", "Cardbox")], count=30
)
with cardboxes:
rep.randomizer.scatter_2d(scatter_plane, check_for_collisions=True)
rep.randomizer.materials(cardbox_mats)
return cardboxes.node
So, for example instead of scattering 30 boxes, I want to scatter a random number uniformly between 10 and 30.