omega-prime-visibility 0.1.0

Omega-Prime-Visibility: Compute Visibility of Objects Polygons

Omega-Prime-Visibility

This packages uses VisiLibity1 through visilibity (LGPL) to comptue the visibility of MovingObjects of omega-prime-Recordings. For every object it computes the visibilty as a float (1.0 for fully visible and 0.0 for not visible) (assuming 2D-brids-eye-view geometries) from the point of view of one object (assumed as a point in the center of the object). In addition the objects occluding the view (occluders) are computed.

This package also provides the computation of visibility as an omega-prime Metric: from omega_prime_visibility import visibility.

The example file highway_merge.mcap is taken from omega-prime and derived from esmini and is under MPL-2.0 license.

License

This python package is distributed under MIT license but the linked libraries visilibity and VisiLibity1 are under the GNU Lesser General Public License (LGPL).

Requirements

You need to have installed swig, boost and cython to be able to install visilibity

On windows you need to additonally install Buildtools:

1. Download Buildtools für Visual Studio 2022
2. Select C++ Tools for Linux Development and install

Installation

pip install omega-prime-visibility

Usage

See ./tutorial.ipynb for detailed instructions.

from omega_prime_visibility import get_visibility_df
import omega_prime
import shapely

r = omega_prime.Recording.from_file('highway_merge.mcap', compute_polygons=True)

obstruction_poly = shapely.Polygon([
    [-220,10],
    [-120,10],
    [-120,-0],
    [-220,-0],
])
df = get_visibility_df(r._df, ego_idx=0, static_occluder_polys=[obstruction_poly])

returns

frame	idx	occluder_idxs	static_occluder_idxs	visibility
0	1	[]	[0]	0.42
0	2	[]	[0]	0
0	3	[]	[0]	0
0	4	[]	[0]	0
0	5	[3]	[0]	0
1	1	[]	[0]	0.53
1	2	[]	[0]	0
1	3	[]	[0]	0
1	4	[]	[0]	0
1	5	[3]	[0]	0
2	1	[]	[0]	0.63
2	2	[]	[0]	0
2	3	[]	[0]	0
2	4	[]	[0]	0
2	5	[3]	[0]	0
3	1	[]	[0]	0.74
...	...	...	...	...
432	3	[]	[]	1
432	4	[1 2 5]	[]	0
432	5	[1 2]	[]	0

Notice

[!IMPORTANT] The project is open-sourced and maintained by the Institute for Automotive Engineering (ika) at RWTH Aachen University. We cover a wide variety of research topics within our Vehicle Intelligence & Automated Driving domain. If you would like to learn more about how we can support your automated driving or robotics efforts, feel free to reach out to us! :email: [e-mail]