SWiM: Spacecraft With Masks (Instance Segmentation)

SWiM: Spacecraft With Masks A large-scale instance segmentation dataset of nearly 64k annotated spacecraft images created using real spacecraft models, superimposed on a mixture of real and synthetic backgrounds generated using NASA’s TTALOS pipeline. To mimic camera distortions and noise in real-world image acquisition, we added different types of noise and distortion. Dataset Summary The dataset contains over 63,917 annotated images with instance masks for varied… See the full description on the dataset page:

Source: Hugging Face Hub (RiceD2KLab/SWiM-SpacecraftWithMasks). Metadata imported from the dataset’s Hub tags.