saeedghsh / Map-Abstraction-2D-GUI

Interactive spatial robot map abstraction with canonical geometric traits.

Map-Abstraction-2D

A pair of GUIs for annotating (auto and manual) spatial maps in Bitmap format with canonical geometric traits (line, circle and their segments). The main underlying representation for these GUIs is the arrangement.

Dependencies and Download

Most dependencies are listed in requirements.txt, and will be installed by the following instructions. But there are two more, namely opencv and arrangement, which should be installed separately.

# Download
git clone https://github.com/saeedghsh/Map-Abstraction-2D.git
cd Map-Abstraction-2D

# Install dependencies
pip install -r requirements.txt

# launch GUIs
python runMe_annotation.py
python runMe_arrangement.py

Annotation GUI

A simple GUI for map annotation with simple geometric traits. This could be useful for fast manual annotation for "ground-truth". The result also could be used by the Arrangement-GUI to creat an arrangement representation of the map. Read on how to use the GUI here.

Arrangement GUI

This GUI is for constructing an arrangement, and visualizing the result. Read on how to use this GUI here.

Laundry List

• (annotation) remove the perpendicular assumption of the automatic dominant orientation.
• (annotation)add automatice trait detection for circle.
• (arrangement) animating the arrangement (?).
• (arrangement) interactive face selection and attribute assignment.
• (arrangement) save arrangement result.
• (arrangement) HowTo document.

These GUIs have been made to facilitate the experimentation and developement of the methods of the following publications:

• S. G. Shahbandi, B. Åstrand and R. Philippsen, "Sensor based adaptive metric-topological cell decomposition method for semantic annotation of structured environments", ICARCV, Singapore, 2014, pp. 1771-1777. doi: 10.1109/ICARCV.2014.7064584 URL.
• S. G. Shahbandi, B. Åstrand and R. Philippsen, "Semi-supervised semantic labeling of adaptive cell decomposition maps in well-structured environments", ECMR, Lincoln, 2015, pp. 1-8. doi: 10.1109/ECMR.2015.7324207 URL.
• S. G. Shahbandi, ‘Semantic Mapping in Warehouses’, Licentiate dissertation, Halmstad University, 2016. URL
• S. G. Shahbandi, M. Magnusson, "2D Map Alignment With Region Decomposition", submitted to Autonomous Robots, 2017.

License

Distributed with a GNU GENERAL PUBLIC LICENSE; see LICENSE.

Copyright (C) Saeed Gholami Shahbandi