AI 용어집
인공지능 완전 사전
Tracking 6DOF
Tracking system that captures the six degrees of freedom (three translations and three rotations) of an object in three-dimensional space. Allows realistic and stable placement of virtual objects relative to the real world.
Point Cloud
Set of points in 3D space used to represent the surface of objects or environments scanned by depth sensors. Forms the basis for 3D reconstruction and surface detection in AR.
Plane Detection
Analysis algorithm that automatically identifies flat surfaces (walls, floors, tables) in the real environment. Essential for allowing users to place virtual objects on real surfaces in a consistent manner.
ARKit
Apple development framework for creating augmented reality applications on iOS and iPadOS. Integrates advanced tracking technologies, surface detection, and lighting estimation.
ARCore
Google development platform that enables the creation of AR applications on Android and iOS. Offers motion tracking, environmental understanding, and light estimation features.
Marker-based AR
AR technology that uses predefined visual markers (images or QR codes) as anchor points to overlay digital content. Simpler to implement but less flexible than markerless approaches.
Markerless AR
Augmented reality approach that does not require predefined markers, instead using object recognition or natural surface detection. Provides a more immersive and spontaneous experience for the user.
Surface Reconstruction
Algorithmic process that reconstructs 3D surfaces from sensor data or 2D images. Fundamental for creating accurate representations of the environment in AR and enabling realistic physical interactions.
Light Estimation
Real-time analysis of the environment's lighting conditions to adapt the appearance of virtual objects. Ensures that augmented content integrates naturally with real shadows and reflections.
Depth Sensing
Capture technology that measures the distance between the sensor and objects in the scene, creating a depth map of the environment. Essential for realistic occlusion and 3D interaction in AR.
Spatial Mapping
Process of creating a 3D representation of the surrounding physical space to enable interaction with virtual objects. Forms the foundation for immersive and contextual AR experiences.
World Tracking
Ability of an AR system to track the position and orientation of the device in world space, independent of initial movements. Enables persistent and accurate placement of virtual objects in the environment.
Image Recognition
Process of identifying and analyzing specific images in the field of view to trigger AR experiences. Enables linking digital content to real-world objects or displays.
Object Recognition
Ability of an AR system to identify and track specific 3D objects in the environment. Differs from image recognition by working with three-dimensional volumes and shapes.
Environmental Understanding
Set of technologies enabling an AR system to analyze and interpret the physical environment in real-time. Includes surface detection, light estimation, and spatial understanding.
Persistent AR
Ability of an AR system to maintain the position and state of virtual objects between different sessions and application restarts. Enables continuous and shared experiences over time.
Cloud Anchoring
Technology that allows saving AR anchor points in the cloud for multi-user and persistent experiences. Facilitates sharing synchronized AR experiences between different devices.