
Image-to-image translation models can be used to mitigate the simulation-to-reality gap in autonomous driving testing, but their effectiveness varies across different tasks, and existing evaluation metrics do not consistently align with the behavior of the autonomous driving systems.


coremsg

evaluating-the-effectiveness-of-image-to-image-translation-models-for-mitigating-the-simulation-to-reality-gap-in-autonomous-driving-testing


Evaluating the Effectiveness of Image-to-Image Translation Models for Mitigating the Simulation-to-Reality Gap in Autonomous Driving Testing
