Glossary Terms

Reliability is the property of consistent intended behaviour and results. Source: IMDRF N67 (2022) Read more

Self-supervised machine learning generates its own labels from otherwise unlabeled data, allowing a model to be trained without manual annotation. Source: FDA Digital Health and AI Glossary Read more

Semi-supervised machine learning combines supervised and unsupervised techniques during training, typically pairing a small amount of labeled data with a much larger body of unlabeled data. Sources: FDA Digital Health and AI Glossary; IMDRF N67 (2022) Read more

SLAM, or simultaneous localization and mapping, is the process by which a device builds a map of its surroundings while at the same time working out its own position within that map. It underpins spatial understanding in augmented reality. Source: Unity Glossary Read more

Spatial audio, also called 3D audio, makes sound appear to originate from specific locations in a virtual environment and responds as the user moves and turns their head. It mimics how sound is perceived in the physical world and contributes substantially to immersion. Sources: Unity Glossary; XRSI Standard XR Definitions (XR-001) Read more

Spatial computing describes interaction with a computer that uses the space around the user as the interface, so digital content is placed and manipulated in three dimensions rather than on a flat screen. Source: XRSI Standard XR Definitions (XR-001) Read more

Spatial immersion is immersion triggered and sustained by the spatial qualities of a virtual environment, such as its sense of place, scale, and surrounding depth. Source: ACM Metaverse Glossary (Zhang, Perkis & Arndt, 2017) Read more

Stereoscopy recreates natural depth perception by presenting a slightly different image to each eye, letting the brain interpret spatial relationships much as it does through binocular vision in the physical world. Source: Unity Glossary Read more

Supervised machine learning trains a model on labeled data, where each example is paired with the correct answer. Sources: FDA Digital Health and AI Glossary; IMDRF N67 (2022) Read more

Synthetic data is created artificially, for example through statistical modeling or computer simulation, so that new values or data elements are generated rather than collected from real subjects. Source: FDA Digital Health and AI Glossary Read more