COCO-Stuff-10K is a dataset designed to enhance scene understanding tasks in computer vision by providing pixel-level annotations for both "things" (discrete objects with well-defined shapes, like cars and people) and "stuff" (amorphous background regions, such as grass and sky). This dataset augments 10,000 images from the original COCO dataset, offering detailed labels across 182 classes—91 "thing" classes and 91 "stuff" classes.
The Common Objects in COntext-stuff (COCO-stuff) dataset is a dataset for scene understanding tasks like semantic segmentation, object detection and image captioning. It is constructed by annotating the original COCO dataset, which originally annotated things while neglecting stuff annotations. There are 164k images in COCO-Stuff-164K dataset that span over 172 categories including 80 things, 91 stuff, and 1 unlabeled class.
The Common Objects in COntext-stuff (COCO-stuff) dataset is a dataset for scene understanding tasks like semantic segmentation, object detection and image captioning. It is constructed by annotating the original COCO dataset, which originally annotated things while neglecting stuff annotations. There are 164k images in COCO-Stuff-164K dataset that span over 172 categories including 80 things, 91 stuff, and 1 unlabeled class.
