Skip to content

Commit

Permalink
update readmes
Browse files Browse the repository at this point in the history
  • Loading branch information
immel-f committed Feb 7, 2024
1 parent 7b19474 commit 57c0231
Show file tree
Hide file tree
Showing 2 changed files with 13 additions and 2 deletions.
2 changes: 1 addition & 1 deletion README.md
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@ https://pypi.org/project/lanelet2/)

Convert Lanelet2 maps directly to instance labels for map perception and other map-based learning tasks! `lanelet2_ml_converter` provides local instance labels of various abstraction layers and representations accessible as numpy arrays from python. Check out the `lanelet2_ml_converter` module for more information and usage examples.

**Note:** As the module is not released officially yet, to try it out you need to either build Lanelet2 with the python bindings yourself or manually install the pip wheel files from the CI artefacts (download [here](https://github.com/immel-f/Lanelet2/actions/runs/7714926040/artifacts/1206339616)).
**Note:** As the module is not released officially yet, to try it out you need to either build Lanelet2 with the python bindings yourself or manually install the pip wheel files from the CI artefacts (download [here](https://github.com/immel-f/Lanelet2/actions/runs/7820288630/artifacts/1228128122)).

## Overview

Expand Down
13 changes: 12 additions & 1 deletion lanelet2_ml_converter/README.md
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,8 @@

Converter module to convert Lanelet2 maps into local instance labels for machine learning tasks.

**Note:** This module is experimental, so the documentation is still sparse and there may be breaking API changes in the future!

![](doc/summary_flowchart.png)

## Usage Examples
Expand Down Expand Up @@ -37,6 +39,15 @@ lanelet::ml_converter::LaneData lData = mDataIf.laneData(True); // get the Lane
TensorData tData = lData.getTensorInstanceData(True, False); // get the local instance labels as Eigen mats
```
## Features
- Access lanelet map information as numpy arrays from python, in a representation directly usable for machine learning tasks
- Compound labels for independence from map annotation artifacts
- Full traceability to the underlying map element for all instances, including the compound instances
- Real-time capable optimized C++ implementation (around 3 ms for one set of local instance labels)
- Save and load generated labels in both binary and human-readable XML format
## Components
### `MapDataInterface`
Expand All @@ -49,7 +60,7 @@ Main interface class that is used to generate `LaneData` objects.
### `TensorInstanceData`
Subclass of `LaneData` that holds all local instance label in a convenient numpy array / Eigen matrix form.
Subclass of `LaneData` that holds all local instance labels in a convenient numpy array / Eigen matrix form.
### `LaneLineStringInstance` / `CompoundLaneLineStringInstance` / `LaneletInstance` / ...
Expand Down

0 comments on commit 57c0231

Please sign in to comment.