The task of identifying and outlining individual objects in an image or video by marking their exact boundaries at the pixel level.
Quality of vision, audio, and image understanding (distinct from modality support)