Pixels of general temporary and permanent objects and obstacles, including mailboxes, fire hydrants, poles, street lights, phone booths, bus stop enclosures, cones, parking meters, etc.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Missing the information I need","missingTheInformationINeed","thumb-down"],["Too complicated / too many steps","tooComplicatedTooManySteps","thumb-down"],["Out of date","outOfDate","thumb-down"],["Samples / code issue","samplesCodeIssue","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-07-14 UTC."],[[["\u003cp\u003e\u003ccode\u003eGARSemanticLabel\u003c/code\u003e defines labels for the Scene Semantics API, enabling detection of various real-world elements in a scene.\u003c/p\u003e\n"],["\u003cp\u003eThe API can identify a range of objects including buildings, trees, vehicles, people, and more, assigning semantic labels to pixels in an image.\u003c/p\u003e\n"],["\u003cp\u003eDevelopers can utilize these labels to understand the composition of a scene and build immersive AR experiences based on the identified elements.\u003c/p\u003e\n"],["\u003cp\u003eEach label represents a specific category like 'sky', 'road', or 'person', with detailed descriptions provided for clarity and accurate interpretation.\u003c/p\u003e\n"],["\u003cp\u003eSome pixels may be categorized as 'unlabeled' if the API cannot confidently assign a semantic label.\u003c/p\u003e\n"]]],["The GARSemanticLabel enumeration defines labels detectable by the Scene Semantics API, mapping them to per-pixel semantic data. This includes labels like Sky, Building, Tree, Road, Sidewalk, Terrain, Structure, Object, Vehicle, Person, and Water. Each label is associated with specific pixel types, such as 'Building' including attached structures, 'Road' for drivable surfaces, and 'Terrain' for walkable vegetation. 'Unlabeled' identifies pixels without semantic labels.\n"],null,["# GARSemanticLabel\n================\n\nSummary\n-------\n\n| ### Enumerations ||\n|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|-------------------------------------------------------------------------------------------------------------------------------|\n| [GARSemanticLabel](#garsemanticlabel)`{` ` `[GARSemanticLabelUnlabeled](#garsemanticlabelunlabeled)` = 0,` ` `[GARSemanticLabelSky](#garsemanticlabelsky)` = 1,` ` `[GARSemanticLabelBuilding](#garsemanticlabelbuilding)` = 2,` ` `[GARSemanticLabelTree](#garsemanticlabeltree)` = 3,` ` `[GARSemanticLabelRoad](#garsemanticlabelroad)` = 4,` ` `[GARSemanticLabelSidewalk](#garsemanticlabelsidewalk)` = 5,` ` `[GARSemanticLabelTerrain](#garsemanticlabelterrain)` = 6,` ` `[GARSemanticLabelStructure](#garsemanticlabelstructure)` = 7,` ` `[GARSemanticLabelObject](#garsemanticlabelobject)` = 8,` ` `[GARSemanticLabelVehicle](#garsemanticlabelvehicle)` = 9,` ` `[GARSemanticLabelPerson](#garsemanticlabelperson)` = 10,` ` `[GARSemanticLabelWater](#garsemanticlabelwater)` = 11` `}` | enum Defines the labels the Scene Semantics API is able to detect and maps human-readable names to per-pixel semantic labels. |\n\nEnumerations\n------------\n\n### GARSemanticLabel\n\n```text\n GARSemanticLabel\n``` \nDefines the labels the Scene Semantics API is able to detect and maps human-readable names to per-pixel semantic labels.\n\nSee [the Scene Semantics Developer Guide](https://developers.google.com/ar/develop/ios/scene-semantics) for more information.\n\n| Properties ||\n|---------------------------|-------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|\n| GARSemanticLabelBuilding | Pixels of buildings, including houses, garages, etc. Includes all structures attached to the building, such as signs, solar panels, scaffolding, etc. |\n| GARSemanticLabelObject | Pixels of general temporary and permanent objects and obstacles, including mailboxes, fire hydrants, poles, street lights, phone booths, bus stop enclosures, cones, parking meters, etc. |\n| GARSemanticLabelPerson | Pixels of humans and animals. |\n| GARSemanticLabelRoad | Pixels of drivable surfaces for vehicles, including paved, unpaved, dirt, driveways, etc. |\n| GARSemanticLabelSidewalk | Pixels of sidewalks for pedestrians and cyclists, including associated curbs. |\n| GARSemanticLabelSky | Pixels of the open sky, including clouds. Thin electrical wires in front of the sky are included, but leaves/vegetation are not included. |\n| GARSemanticLabelStructure | Pixels of structures that are not buildings, including fences, guardrails, stand-alone walls, tunnels, bridges, mountains, etc. |\n| GARSemanticLabelTerrain | Pixels of walkable vegetation areas, including grass, soil, sand, etc. In contrast, 'tree' specifies non-walkable vegetation, like trees and bushes. |\n| GARSemanticLabelTree | Pixels of non-walkable vegetation, like trees and shrubs. In contrast, 'terrain' specifies walkable vegetation, like grass. |\n| GARSemanticLabelUnlabeled | Pixels with no semantic label available in the API output. |\n| GARSemanticLabelVehicle | Pixels of vehicles, including cars, vans, buses, trucks, motorcycles, bicycles, trains, etc. |\n| GARSemanticLabelWater | Pixels of ground surfaces covered by water, including lakes, rivers, etc. |"]]
