Semantic Scene Graph

A semantic scene graph is a structured data representation of a 3D or 2D environment that encodes objects, their properties (such as position, size, color, and audio characteristics), and the spatial and hierarchical relationships between them. In accessibility research, semantic scene graphs enable AI systems to reason about and describe environments to blind and low-vision users, supporting queries like identifying the nearest object, describing relative positions in egocentric terms (in front of, to the left), or modifying object properties on demand. When augmented with accessibility-specific metadata such as egocentric spatial relations and multimodal attributes, semantic scene graphs provide the contextual grounding needed for reliable natural language accessibility interactions.