FraunhoferIOSB/SynsetSignsetSensitivityAnalysisDataset

Synset Signset Sensitivity Analysis Dataset The Synset Signset Sensitivity Analysis Dataset is a large-scale, physically based synthetic test dataset for evaluating the sensitivity and robustness of machine-learning traffic sign recognition models to sensing parameters. It contains 9,442,800 images from 43 GTSRB traffic sign classes, systematically generated across 2,196 combinations of motion blur and additive image noise. The dataset also includes traffic sign masks… See the full description on the dataset page: https://huggingface.co/datasets/FraunhoferIOSB/SynsetSignsetSensitivityAnalysisDataset.