Releases: isl-org/Open3D-ML
Releases · isl-org/Open3D-ML
v0.18 release
What's Changed
- Fix pytorch 1.13 error due to 2D weights by @ssheorey in #601
- Avoid renaming unwanted part of the path in dataset loading scripts by @ManonCortial in #597
- corrected pairing in SSD head of pointpillars by @zhongyidu in #602
- Replace colon with dash in timestamp variables by @Cassius0924 in #620
- Support python 311 by @ssheorey in #619
- master -> main for branches by @ssheorey in #626
- Add Pandaset class by @SulekBartek in #611
- Add tum facade dataset by @ThomasFroech in #613
New Contributors
- @ManonCortial made their first contribution in #597
- @zhongyidu made their first contribution in #602
- @Cassius0924 made their first contribution in #620
- @SulekBartek made their first contribution in #611
- @ThomasFroech made their first contribution in #613
Full Changelog: v0.17.0...v0.18.0
Contributors
ssheorey, ThomasFroech, and 4 other contributors
Assets 2
v0.17 release
What's Changed
- fixed 403 by @Birkenpapier in #575
- Update RandLANet weights link by @friendship1 in #579
- Upgrade PyTorch to 1.13.1 by @ssheorey in #583
- RandLaNet update_probs label smoothing logic by @mariusud in #569
- Dev to master 0.17 by @ssheorey in #586
New Contributors
- @Birkenpapier made their first contribution in #575
- @friendship1 made their first contribution in #579
- @mariusud made their first contribution in #569
Full Changelog: v0.16.0...v0.17.0
Contributors
Birkenpapier, friendship1, and 2 other contributors
Assets 2
v0.16.0
v0.15.1
Merge pull request #477 from isl-org/ssheorey/docs-fix Fix RandLANet docs
v0.14.1
Downgrade torch on mac to 1.8.1 (#436) * Downgrade torch on mac to 1.8.1 * Update CI to use old CXX11_ABI due to new CXX11_ABI default in Open3D. Co-authored-by: Sameer Sheorey <[email protected]>
v0.13.1
r0.13.1 fix import issue (#290)
v0.13.0
Update download link (#288) Fixed RandLANet mIoU issue on S3DIS dataset and tensorflow. Updated weights link.
Open3D-ML 0.12.0
Check out all the new features here
- New 3D object detection module
- Training and inference pipelines in TensorFlow and PyTorch
- 3D BoundingBox visualizer
- PointPillars
- Datasets: KITTI, Waymo, Lyft level 5, nuSecenes, and Argoverse.