Research papers · Posted 2022.01.10 · Publication / acceptance: author-reported
Vision in adverse weather: Augmentation using CycleGANs with various object detectors for robust perception in autonomous racing
Vision in adverse weather: Augmentation using CycleGANs with various object detectors for robust perception in autonomous racing
Overview
The study synthesizes adverse-weather racing images with CycleGAN and uses them for detector training. Comparisons across object detectors and dataset combinations examine robustness under nighttime conditions and lens droplets.
Publication and version information
Publication, acceptance or venue information is author-reported on arXiv. The exact source wording is preserved below.
Author-supplied source information
Author-supplied information on arXiv.
ICML 2022
Version: arXiv v3 — manuscript revision