This dataset contains labeled images of vortex structures across physical scales, from atmospheric phenomena (tornadoes, waterspouts, jellyfish clouds) to astronomical objects (nebular pillars, proplyds, filaments). The goal is to train a neural network that identifies vortex morphologies (columns, rings, braids, Y-junctions, X-crossings… See the full description on the dataset page: https://huggingface.co/datasets/JimGalasyn/vortex-vision-data.