Semantic Segmentation With 171 Extended Object Stuff Categories Via Coco Stuff Variant

via “semantic segmentation with 171 extended object/stuff categories via coco-stuff variant”

330K images with object detection, segmentation, and captions.

Unique: 171-category taxonomy combining 80 instance objects + 91 stuff categories enables panoptic segmentation in single dataset; pixel-level masks for stuff enable dense scene understanding without instance boundaries

vs others: More comprehensive than ADE20K (150 categories) and larger scale than Cityscapes (5K images); unified instance+stuff annotation enables panoptic evaluation unlike separate semantic/instance datasets