A technique that divides an image into small rectangular sections (patches) and processes each section separately to extract visual features more efficiently.
Quality of vision, audio, and image understanding (distinct from modality support)