docs(readme): show Patchwork++ as ROS 2 default + advertise SemanticKITTI eval harness#98
Merged
Merged
Conversation
…harness ROS 2 section: - Make explicit that `ros2 launch patchworkpp patchworkpp.launch.py` alone runs Patchwork++; only `algorithm:=patchwork` opts into the classic algorithm. The launch file's LaunchConfiguration default is already "patchworkpp" (ros/launch/patchworkpp.launch.py:20), so this is a doc fix, not a behavioural change. What's-in-this-repo: - Add a bullet for python/examples/evaluate_semantickitti.py and note that the two algorithms use different SemanticKITTI evaluation protocols (Patchwork paper counts low-z vegetation as ground; Patchwork++ paper Sec. IV.A excludes vegetation entirely because the SemanticKITTI `vegetation` label mixes spurious leaves / branches with low ground cover). Point at USAGE.md §1 and §4 for the protocol switch and per-sequence numbers. Docs-only; no version bump.
…ntributors note pypatchworkpp ships on PyPI since v1.3.0; the pip banner is already at the top of the README, but the Python install section was still making `make pyinstall` the headline command. Switch the order: - Lead with `pip install pypatchworkpp` and `pip install 'pypatchworkpp[demo]'` (the `[demo]` extra is defined in python/pyproject.toml and pulls open3d). - Keep `make pyinstall` / `make pyinstall_with_demo` in a collapsed <details> block labelled "Build from source (contributors / unreleased main)" so the source-build path is still discoverable without competing with the pip path for new users. Docs-only; same PR as the ROS 2 default clarification + eval-harness bullet.
The README "What's in this repo" bullet was carrying a 4-line wall of text about why Patchwork and Patchwork++ disagree on the vegetation class. Shrink it to a single sentence + link to USAGE.md §1, and put the full rationale (low ground cover vs. overhead foliage, why the two papers chose different resolutions) where it belongs.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Two small README clarifications.
1. ROS 2 default is Patchwork++
The README previously showed only
ros2 launch patchworkpp patchworkpp.launch.py algorithm:=patchwork, which reads as if classic Patchwork is the default. The launch file atros/launch/patchworkpp.launch.py:20and the C++ default atros/src/GroundSegmentationServer.cpp:95are already"patchworkpp", so this is a doc fix. New text shows both invocations explicitly:2. Advertise the SemanticKITTI eval harness + protocol caveat
Add a bullet to "What's in this repo" pointing at
python/examples/evaluate_semantickitti.pyand noting that the two algorithms use different evaluation protocols — the original Patchwork paper counts low-z vegetation as ground while the Patchwork++ paper (Sec. IV.A) excludes vegetation entirely, because the SemanticKITTIvegetationlabel mixes spurious leaves / branches with low ground cover and a single ground/non-ground choice is impractical. Points users at USAGE.md §1 and §4 for the protocol switch and per-sequence numbers.Test plan
pre-commit runclean (mdformat applied one auto-fix on the new bullet line, then green).patchworkpp.evaluate_semantickitti.py).Docs-only; no version bump.