README.md.bak

# :cityscape: TUM - FACADE :cityscape:
**Benchmark for semantic facade segmentation on MLS point clouds** 

![](https://github.com/OloOcki/tum-facade/blob/main/img/intro.gif)

## :star2: Highlights :star2:

- MLS point clouds with **facade-level labels** (i.e., windows, doors, balconies, moldings, etc.)
- **14 annotated facades** and 15 non-annotated for further benchmark extension or testing
- total no of annotated points
- in **local** and **global** coordinate reference system
- settings file for **adding your own data**

## :mag_right: Available labels

<p float="center">
    <img src="documentation/img/classesTableHex.png" width="49%" title="Available classes table"/>
    <img src="documentation/images/bldID62.png" width="49%" title="bld id 62"/>
</p>

-download button

## :bar_chart: Statistics

- no fo annotated points
- bar (?) chart with data distribution

## :construction_worker: Settings file and labeling process

-link to settings
-link to Hitachi

## :mortar_board: Paper

The introduction of this dataset and overview of the point clouds benchmark landscape are available here:

```plain
@article{wysocki,
  title = {...},
  author = {...},
  journal = {...},
  year = {2022},
  month = ...,
  volume = {...},
  number = {...},
  pages = {...},
  publisher = {...},
  doi = {...},
  url = {...}
}
```

## :handshake: Acknowledgments 

-Jiarui Zhang
-TUM-MLS dataset
-Hitachi labeling tool