Object Detection
TensorRT
ONNX
autoware
lidar
point-cloud
3d-object-detection
centerpoint
pointpillars
ros2
autonomous-driving
Instructions to use AutowareFoundation/lidar_centerpoint with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- TensorRT
How to use AutowareFoundation/lidar_centerpoint with TensorRT:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Refactor to per-variant folders (base/tiny/sigma/short_range), uniform filenames
#1
by Technolojin - opened
Restructures the bundle into self-contained per-variant folders with uniform filenames:
base/ tiny/ sigma/ short_range/
βββ pts_voxel_encoder.onnx
βββ pts_backbone_neck_head.onnx
βββ ml_package.param.yaml
βββ detection_class_remapper.param.yaml
βββ deploy_metadata.yaml
sigmaONNX weights are published for the first time (previously the repo carried only its parameter file).short_rangeis added as a fourth variant (previously not distributed at all); itsml_package.param.yamlgains thedetection_score_thresholdsblock required by the current node.- Flat root-level files are removed; the previous layout remains available via the
v3.0tag.
Companion PRs (to follow): autoware_universe (launch/config), autoware_launch (launcher + mirrors), autoware (ansible revision bump to v4.0).
Maintainer action after merge: please create tag v4.0 on the merge commit β the ansible artifacts role pins --revision v4.0.
Technolojin changed pull request status to open
refactoring ongoing. please wait until done.
ready for review. please see https://github.com/autowarefoundation/autoware_universe/issues/13086 for overall task.
I tested all the models, fixed the sigma variant.