Overview
ModaNet is a dataset of street fashion images annotated with polygons, designed to serve as a benchmark for computer vision research. It provides multiple polygon annotations for each image, with each polygon associated with a label from 13 meta fashion categories. The annotations are based on images in the PaperDoll image set but extend it significantly. ModaNet supports tasks such as semantic segmentation, object detection, and instance segmentation, providing leaderboards in COCO-style through EvalAI. The dataset's format follows the COCO-dataset style. It is intended to facilitate research in areas including polygon detection and segmentation.
