Fetoscopy Placenta Dataset

First Fetoscope Placenta Dataset for In-Vivo Vessel Segmentation and MosaickingOverviewThe fetoscope placenta dataset described in [Bano et al. MICCAI2020] consists of in-vivo fetoscopic imagery from twin-to-twin transfusion syndrome (TTTS) procedures. It includes:483 frames from 6 fetoscopic videos with expert ground-truth placental vessel segmentation annotations.6 additional continuous fetoscopic video clips (total 950 frames) without manual labels but accompanied by predicted vessel segmentation maps generated via leave-one-out cross-validation.The dataset is intended for research on placental vessel segmentation and fetoscopic video mosaicking/registration under realistic surgical conditions (limited field of view, specularities, and variable visibility).Citing the DatasetCite [Bano et al. MICCAI2020] whenever research making use of this dataset is reported in any academic publication or research report.