PlateShapez

PlateShapez is a research tool designed to create adversarially perturbed license plate overlays on vehicle images, facilitating the generation of structured datasets while ensuring reproducibility and ethical considerations. The tool offers a user-friendly command-line interface and a Python API to customize overlay perturbations, such as shapes, noise, and textures, allowing for in-depth exploration of dataset creation options. Notable features include customizable configuration files, interactive demos, and robust logging with metadata saving capabilities.