Artifacts for the paper "Concretization of Abstract Traffic Scene Specifications Using Multi-objective Optimization"

This deposit contains measurement data and additional artifacts pertaining to the "Concretization of Abstract Traffic Scene Specifications Using Multi-objective Optimization" paper. Specifically, the artifacts are divided into 5 directories: <em>0-config/ </em>contains configuration files use to generate input scene specifications (using Scenic) with various number of actors. <em>1-data/</em> contains the generated input scene specifications for each map and actor size. It also contains additional scene specifications used for RQ2 (in the `zalaFullcrop/` subfolder). <em>2-results/</em> contains measurement results (i.e. (1) visualizations of generated scenes and (2) corresponding scene specifications containing exact positions of each vehicle, as well as (3) measurement data) for each research question. <em>3-figures/</em> contains figures derived from the contents of `results/` (including additional figures not included in the publication). <em>4-TestingSemanticSegmentation/</em> contains artifacts related to the integration of the CARLA simulator and of three computer-vision components: <em> 0-sceneConfig/</em> contains configuration files used to generate scenes. <em>1-scenesWithExactCoordinates/ </em>contains (1) visualizations of generated scenes and (2) corresponding Scenic-compatible scene specifications containing exact positions of each vehicle. <em> 2-staticImages/</em> contains dashcam images for generated scenes and corresponding ground truth semantic segmentation, obtained through the CARLA integration. <em>3-predictions/</em> contains predicted semantic segmentation results (and their overlay on top of the corresponding dashcam image) obtained using three computer-vision components. <em>4-videos/</em> contains videos of generated scenes running with the default AV stack included in CARLA. <em>5-metrics</em> contains initial measurement data for the predicted semantic segmentation. All the code of the proposed MOO-based scene concretization approach is implemented as an extension to the Scenic tool, which is available under the 3-Clause BSD License.