Spatial computing: Difference between revisions
Xinreality (talk | contribs) No edit summary |
Xinreality (talk | contribs) m Text replacement - "aiwiki:Artificial Intelligence" to "artificial intelligence" |
||
| (One intermediate revision by the same user not shown) | |||
| Line 23: | Line 23: | ||
Spatial computing typically involves several key components working together: | Spatial computing typically involves several key components working together: | ||
* '''Machine Perception of Space:''' Devices must understand the physical environment in 3D. This involves technologies like [[Simultaneous Localization and Mapping]] (SLAM) to track the device's position and orientation while building a map of the space.<ref name="DurrantWhyteSLAM"/> [[Depth sensor]]s (like [[LiDAR]] or Time-of-Flight cameras) and [[RGB camera]]s capture geometric and visual information. [[Computer vision]] algorithms, often powered by [[artificial intelligence]] (AI), interpret this data to recognize surfaces, objects (for example walls, tables, chairs), people, and potentially understand scene semantics.<ref name="CogentSLAM"/><ref name="TechTargetWhatIs"/> | * '''Machine Perception of Space:''' Devices must understand the physical environment in 3D. This involves technologies like [[Simultaneous Localization and Mapping]] (SLAM) to track the device's position and orientation while building a map of the space.<ref name="DurrantWhyteSLAM"/> [[Depth sensor]]s (like [[LiDAR]] or Time-of-Flight cameras) and [[RGB camera]]s capture geometric and visual information. [[Computer vision]] algorithms, often powered by [[aiwiki:Artificial Intelligence|artificial intelligence]] (AI), interpret this data to recognize surfaces, objects (for example walls, tables, chairs), people, and potentially understand scene semantics.<ref name="CogentSLAM"/><ref name="TechTargetWhatIs"/> | ||
* '''Persistence and Context:''' Digital objects or information placed within the spatial environment can maintain their position and state relative to the physical world, even when the user looks away or leaves and returns (spatial anchors). The system uses its understanding of spatial context to anchor digital elements appropriately and realistically, potentially enabling occlusion (virtual objects appearing behind real ones) and physics interactions.<ref name="HandwikiHistory"/> | * '''Persistence and Context:''' Digital objects or information placed within the spatial environment can maintain their position and state relative to the physical world, even when the user looks away or leaves and returns (spatial anchors). The system uses its understanding of spatial context to anchor digital elements appropriately and realistically, potentially enabling occlusion (virtual objects appearing behind real ones) and physics interactions.<ref name="HandwikiHistory"/> | ||
* '''Natural User Interaction:''' Input moves beyond the [[keyboard]] and [[mouse]]. Common interaction methods include [[Hand tracking]] (recognizing hand shapes and gestures), [[Eye tracking]] (using gaze as a pointer or input trigger), [[Voice command]]s, and sometimes specialized controllers. The goal is intuitive interaction that mimics how humans interact with the physical world, making the computer interface feel "invisible."<ref name="PCMagWhatIs"/><ref name="Microsoft HoloLens"/> | * '''Natural User Interaction:''' Input moves beyond the [[keyboard]] and [[mouse]]. Common interaction methods include [[Hand tracking]] (recognizing hand shapes and gestures), [[Eye tracking]] (using gaze as a pointer or input trigger), [[Voice command]]s, and sometimes specialized controllers. The goal is intuitive interaction that mimics how humans interact with the physical world, making the computer interface feel "invisible."<ref name="PCMagWhatIs"/><ref name="Microsoft HoloLens"/> | ||