In order to allow the development of reliable software components for autonomous driving, we have curated a novel dataset targeted primarily for performance evaluation. The dataset has two subsets targeted for two different use cases, e.g. Tram Railway segmentation and Traffic Sign Detection/Identification. This result is further described on the project website (https://sites.google.com/view/didymos-tacr/results). The raw data were collected privately or provided by Plzeňské Městské dopravní podniky (PMDP).

Článek představuje analýzu datových korpusů z oblasti pokročilých systémů asistence řidiče. Věnujeme se kvalitativním vlastnostem existujících datasetů, představujeme problémy, které stávajíci datasety mají, navrhujeme jejich řešení a představujeme novou metriku kvality detekce objektů zavislou na rychlosti vozidla.

As part of the DiDYMOS project, a 3D model of the city of Plzeň was created and expanded with data layers focused on supporting autonomous driving. Key activities included mobile mapping of the test section, segmentation and extraction of geographic features from the point cloud, and their integration into the model. The main benefits were the high accuracy and detail of the data obtained, which enabled the mapping of technical and transport infrastructure, including rails, curbs, traffic signs, and other vertical and surface features. The data were processed in various formats and used not only for autonomous driving but also for broader urban administration and planning needs.