ISO/IEC NP 26073

Spatio-Temporal Mixed and Augmented Reality Experience Description (MAR-ED) for Interactive Playback

Апстракт

This standard specifies the Spatio-Temporal Mixed and Augmented Reality Experience Description (MAR-ED), a framework for representing past events to enable interactive and adaptive playback within a user's present physical space. The framework's purpose is to standardize how experiences are described not as a collection of static visual data, but as a structured narrative of meaningful events, interactions, and spatial relationships. This approach transforms a canonical recorded experience into a unique adaptive MAR experience, where its spatio-temporal structure dynamically conforms to a new environment and its narrative can be altered by live user input.

This draft defines the three core sets of descriptive primitives that constitute the MAR-ED framework. It details the Event Primitives (SemanticExperienceSegment, InteractionEvent, StateChangeEvent) used to transform raw spatio-temporal data into a structured semantic log. It also defines the Keyframe Primitives (KeyInteraction, KeyStateChange, KeyframeDecisionThreshold) that provide a mechanism to distill the continuous flow of events into a discrete set of semantically significant keyframes for efficient access.

Furthermore, this draft covers the Playback Primitives, which provide the runtime logic for user-driven adaptive playback. This includes the principles for TimelineAdaptation to adjust the experience's pacing, CreateNewBranch to manage deviations from the recorded narrative in response to user interaction, and ReturnToMain to ensure a seamless transition back to the core narrative. Applications of this framework are presented to illustrate how recorded expert tutorials, cultural heritage events, or personal memories can evolve beyond passive videos into immersive, spatially-integrated experiences.

Overall, this draft addresses the concepts, semantics, and descriptive model for the MAR-ED framework and specifies:

a descriptive model for representing spatio-temporal experiences based on semantic events rather than raw data,

data primitives and their classes that constitute the model, covering events, keyframes, and playback logic for user-driven adaptation and interaction,

the runtime logic that supports the generation of new narrative branches in response to live user input and the seamless return to the original timeline,

Use cases and examples demonstrating how recorded spatio-temporal data can be transformed into dynamic, personalized, and adaptive MAR experiences in new contexts.