Data preprocessing & augmentation framework, designed for working with crowd counting datasets, ML/DL framework-independent. Supports multitude of simple as well as advanced transformations, outputs and loaders, all of them to be combined using pipelines.
