A
Aliasing
Visual artifacts (e.g. jagged edges or moiré patterns) that occur when a continuous signal is undersampled.
Aliasing
Visual artifacts (e.g. jagged edges or moiré patterns) that occur when a continuous signal is undersampled.
Alpha Blending
A compositing technique that combines overlapping 3D Gaussians based on their transparency (alpha values) to render smooth and accurate images.
Alpha Blending
A compositing technique that combines overlapping 3D Gaussians based on their transparency (alpha values) to render smooth and accurate images.
Ambient Occlusion
A shading method that darkens areas with little ambient light exposure (e.g. creases and corners).
Ambient Occlusion
A shading method that darkens areas with little ambient light exposure (e.g. creases and corners).
Anti-Aliasing
Techniques used to smooth out jagged edges and reduce visual artifacts caused by undersampling.
Anti-Aliasing
Techniques used to smooth out jagged edges and reduce visual artifacts caused by undersampling.
Attenuation
The gradual reduction of light intensity as it travels through a medium
Attenuation
The gradual reduction of light intensity as it travels through a medium
B
Backpropagation
An algorithm that computes gradients for all network weights by propagating errors backward from the output.
Backpropagation
An algorithm that computes gradients for all network weights by propagating errors backward from the output.
Bidirectional Reflectance Distribution Function (BRDF)
A function describing how light is reflected at a surface from an incoming to an outgoing direction.
Bidirectional Reflectance Distribution Function (BRDF)
A function describing how light is reflected at a surface from an incoming to an outgoing direction.
Bidirectional Scattering Distribution Function (BSDF)
A function that models both reflected and transmitted light at a surface.
Bidirectional Scattering Distribution Function (BSDF)
A function that models both reflected and transmitted light at a surface.
Bounding Volume Hierarchy (BVH)
A tree structure that groups scene elements in nested bounding boxes to accelerate ray–object intersection tests.
Bounding Volume Hierarchy (BVH)
A tree structure that groups scene elements in nested bounding boxes to accelerate ray–object intersection tests.
Bundle Adjustment
An optimization technique that refines camera poses and 3D structure by minimizing reprojection errors.
Bundle Adjustment
An optimization technique that refines camera poses and 3D structure by minimizing reprojection errors.
C
Camera Calibration
The process of determining a camera’s intrinsic (lens) and extrinsic (position/orientation) parameters.
Camera Calibration
The process of determining a camera’s intrinsic (lens) and extrinsic (position/orientation) parameters.
Camera Extrinsics
The parameters (position and orientation) that describe a camera’s placement in the world.
Camera Extrinsics
The parameters (position and orientation) that describe a camera’s placement in the world.
Camera Intrinsics
The internal parameters of a camera (e.g. focal length, sensor size, and lens distortion).
Camera Intrinsics
The internal parameters of a camera (e.g. focal length, sensor size, and lens distortion).
Camera Pose
The position and orientation of a camera within a 3D scene.
Camera Pose
The position and orientation of a camera within a 3D scene.
COLMAP
A popular open-source tool for Structure from Motion (SfM) and Multi-View Stereo (MVS). It processes photos to generate 3D reconstructions, often serving as the starting point for Radiance Field reconstructions.
COLMAP
A popular open-source tool for Structure from Motion (SfM) and Multi-View Stereo (MVS). It processes photos to generate 3D reconstructions, often serving as the starting point for Radiance Field reconstructions.
Computer Graphics
The field focused on generating and manipulating images and visual content using computers.
Computer Graphics
The field focused on generating and manipulating images and visual content using computers.
Computer Vision
Artificial intelligence (AI) that enables computers to see, analyze, and understand the visual world.
Computer Vision
Artificial intelligence (AI) that enables computers to see, analyze, and understand the visual world.
D
Deferred Neural Rendering
Definition: A rendering pipeline that combines traditional computer graphics techniques with neural networks to produce high-fidelity images.
Context: Enhances rendering quality and efficiency by applying neural networks to intermediate representations of a scene.
Deferred Neural Rendering
Definition: A rendering pipeline that combines traditional computer graphics techniques with neural networks to produce high-fidelity images.
Context: Enhances rendering quality and efficiency by applying neural networks to intermediate representations of a scene.
Deferred Rendering
A two-pass rendering technique that first collects geometric data and then applies lighting in screen space.
Deferred Rendering
A two-pass rendering technique that first collects geometric data and then applies lighting in screen space.
Densification
Densification in the context of Gaussian splatting involves adding new Gaussian primitives to enhance the representation of a scene, particularly in regions lacking detail. This process aims to improve image quality by populating sparse areas with additional Gaussians.
Densification
Densification in the context of Gaussian splatting involves adding new Gaussian primitives to enhance the representation of a scene, particularly in regions lacking detail. This process aims to improve image quality by populating sparse areas with additional Gaussians.
Density (Volumetric)
A scalar field that indicates the opacity or concentration of matter at a point in a volume.
Density (Volumetric)
A scalar field that indicates the opacity or concentration of matter at a point in a volume.
Depth Map
An image in which each pixel’s value represents the distance from the camera to the nearest surface.
Depth Map
An image in which each pixel’s value represents the distance from the camera to the nearest surface.
Differentiable Rendering
Definition: A rendering approach that allows gradients of images with respect to scene parameters to be computed, facilitating the integration of rendering into machine learning pipelines.
Context: Enables optimization and learning of scene properties directly from images.
Differentiable Rendering
Definition: A rendering approach that allows gradients of images with respect to scene parameters to be computed, facilitating the integration of rendering into machine learning pipelines.
Context: Enables optimization and learning of scene properties directly from images.
Diffuse Reflection
Light reflection that scatters equally in all directions.
Diffuse Reflection
Light reflection that scatters equally in all directions.
Dynamic NeRF
A neural radiance field that includes time as an input.
Dynamic NeRF
A neural radiance field that includes time as an input.
E
Emission
Light that is produced by a surface or volume.
Emission
Light that is produced by a surface or volume.
Environment Map
A panoramic image that represents the color and intensity of light coming from all directions in a scene.
Environment Map
A panoramic image that represents the color and intensity of light coming from all directions in a scene.
Epipolar Geometry
The geometric relationship between two camera views that constrains where corresponding points can appear.
Epipolar Geometry
The geometric relationship between two camera views that constrains where corresponding points can appear.
Explicit Representation
A scene model that directly stores geometric primitives (e.g. meshes or point clouds) rather than encoding them implicitly. Gaussian Splatting has an explicit representation.
Explicit Representation
A scene model that directly stores geometric primitives (e.g. meshes or point clouds) rather than encoding them implicitly. Gaussian Splatting has an explicit representation.
F
Field of View (FOV)
The angular extent of the scene that a camera captures.
Field of View (FOV)
The angular extent of the scene that a camera captures.
Fourier Feature Mapping
Definition: A technique that maps input coordinates into a higher-dimensional space using sinusoidal functions to enable neural networks to learn high-frequency variations in data.
Context: Addresses the spectral bias of neural networks, improving their ability to represent detailed structures.
Fourier Feature Mapping
Definition: A technique that maps input coordinates into a higher-dimensional space using sinusoidal functions to enable neural networks to learn high-frequency variations in data.
Context: Addresses the spectral bias of neural networks, improving their ability to represent detailed structures.
Fourier Feature Mapping (Positional Encoding)
A technique that transforms spatial coordinates into high-dimensional sine and cosine features to capture fine details.
Fourier Feature Mapping (Positional Encoding)
A technique that transforms spatial coordinates into high-dimensional sine and cosine features to capture fine details.
Free Viewpoint Video
Video generated from a scene representation that allows interactive changes of the viewing perspective.
Free Viewpoint Video
Video generated from a scene representation that allows interactive changes of the viewing perspective.
G
Gaussian Splatting (3DGS)
A radiance field representation that uses a sparse cloud of 3D Gaussians, enabling real-time rendering of radiance fields.
Gaussian Splatting (3DGS)
A radiance field representation that uses a sparse cloud of 3D Gaussians, enabling real-time rendering of radiance fields.
Generalization
A model’s ability to perform well on new unseen data beyond its training examples.
Generalization
A model’s ability to perform well on new unseen data beyond its training examples.
Global Illumination
A set of techniques that simulate both direct and indirect lighting to achieve realistic image synthesis.
Global Illumination
A set of techniques that simulate both direct and indirect lighting to achieve realistic image synthesis.
Gradient Descent
An iterative optimization algorithm that updates model parameters in the direction that most reduces error.
Gradient Descent
An iterative optimization algorithm that updates model parameters in the direction that most reduces error.
Graphics Pipeline
The sequence of stages (e.g. transformation, rasterization, and shading) that convert 3D scene data into a 2D image.
Graphics Pipeline
The sequence of stages (e.g. transformation, rasterization, and shading) that convert 3D scene data into a 2D image.
Graphics Processing Unity (GPU)
A specialized processor optimized for parallel computations used extensively in rendering.
Graphics Processing Unity (GPU)
A specialized processor optimized for parallel computations used extensively in rendering.
I
Image-Based Rendering (IBR)
Techniques that synthesize new views directly from captured images without explicitly reconstructing full 3D geometry.
Image-Based Rendering (IBR)
Techniques that synthesize new views directly from captured images without explicitly reconstructing full 3D geometry.
Implicit Representation
A scene model defined by a continuous function (often via neural networks) that can be queried for properties like color and density. NeRFs use an Implicit Representation.
Implicit Representation
A scene model defined by a continuous function (often via neural networks) that can be queried for properties like color and density. NeRFs use an Implicit Representation.
Implicit Surface Reconstruction
Definition: A mathematical description of a surface defined by an equation, where the surface consists of all points satisfying the equation.
Context: Used in computer graphics to model complex shapes and surfaces without explicitly defining their geometry.
Implicit Surface Reconstruction
Definition: A mathematical description of a surface defined by an equation, where the surface consists of all points satisfying the equation.
Context: Used in computer graphics to model complex shapes and surfaces without explicitly defining their geometry.
Index of Refraction
A property of a medium that quantifies how much it bends light as the light passes through it.
Index of Refraction
A property of a medium that quantifies how much it bends light as the light passes through it.
Instant NeRF
A fast neural radiance field approach, also known as Instant NGP that uses techniques such as hash encoding to dramatically reduce training time.
Instant NeRF
A fast neural radiance field approach, also known as Instant NGP that uses techniques such as hash encoding to dramatically reduce training time.
Inverse Rendering
Definition: A technique aimed at determining the physical properties of a scene (such as geometry, materials, and lighting) from images.
Context: Enables the creation of editable 3D models from photographs, facilitating tasks like object recognition and scene understanding.
Inverse Rendering
Definition: A technique aimed at determining the physical properties of a scene (such as geometry, materials, and lighting) from images.
Context: Enables the creation of editable 3D models from photographs, facilitating tasks like object recognition and scene understanding.
L
LiDAR
A sensor technology that uses laser pulses to measure distances and generate a 3D point cloud of an environment.
LiDAR
A sensor technology that uses laser pulses to measure distances and generate a 3D point cloud of an environment.
Light Field
A function that describes the intensity and direction of light rays throughout a scene.
Light Field
A function that describes the intensity and direction of light rays throughout a scene.
M
Machine Learning
A type of artificial intelligence that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.
Machine Learning
A type of artificial intelligence that provides systems the ability to automatically learn and improve from experience without being explicitly programmed.
Material
the optical properties (e.g. color and reflectivity) that determine how a surface interacts with light.
Material
the optical properties (e.g. color and reflectivity) that determine how a surface interacts with light.
Mesh
A collection of vertices, edges, and faces that explicitly define the surface geometry of a 3D object.
Mesh
A collection of vertices, edges, and faces that explicitly define the surface geometry of a 3D object.
Mip-NeRF
A variant of NeRF that integrates over a ray’s conical frustum (similar to mipmapping) to reduce aliasing.
Mip-NeRF
A variant of NeRF that integrates over a ray’s conical frustum (similar to mipmapping) to reduce aliasing.
Monte Carlo Rendering
A rendering method that uses random sampling (e.g. path tracing) to solve the rendering equation.
Monte Carlo Rendering
A rendering method that uses random sampling (e.g. path tracing) to solve the rendering equation.
Multi-Layer Perceptron (MLP)
A fully connected neural network that maps inputs (such as 3D coordinates) to outputs (like color and density). Often used in NeRFs.
Multi-Layer Perceptron (MLP)
A fully connected neural network that maps inputs (such as 3D coordinates) to outputs (like color and density). Often used in NeRFs.
Multi-View Stereo (MVS)
Techniques for reconstructing dense 3D models from multiple overlapping images.
Multi-View Stereo (MVS)
Techniques for reconstructing dense 3D models from multiple overlapping images.
N
Neural Network
A computational model composed of interconnected layers of nodes that learns complex functions from data.
Neural Network
A computational model composed of interconnected layers of nodes that learns complex functions from data.
Neural Radiance Fields (NeRF)
A Radiance Field representation that reconstructs lifelike 3D scenes from 2D images by modeling the color and density of a scene.
Neural Radiance Fields (NeRF)
A Radiance Field representation that reconstructs lifelike 3D scenes from 2D images by modeling the color and density of a scene.
Neural Rendering
A technique that combines deep learning models with traditional computer graphics to generate, reconstruct, or enhance images and videos.
Neural Rendering
A technique that combines deep learning models with traditional computer graphics to generate, reconstruct, or enhance images and videos.
Novel View Synthesis
The process of generating novel images of a scene from new viewpoints, based on a set of existing images.
Context: Applications include virtual reality, 3D reconstruction, and image-based rendering.
Novel View Synthesis
The process of generating novel images of a scene from new viewpoints, based on a set of existing images.
Context: Applications include virtual reality, 3D reconstruction, and image-based rendering.
O
Occlusion
The blocking of light or a view by an object resulting in hidden or shadowed areas.
Occlusion
The blocking of light or a view by an object resulting in hidden or shadowed areas.
Opacity
A measure of how impenetrable an object is to light; the opposite of transparency.
Opacity
A measure of how impenetrable an object is to light; the opposite of transparency.
Optical Depth
A measure of the cumulative opacity of a medium along a ray calculated as the integral of density.
Optical Depth
A measure of the cumulative opacity of a medium along a ray calculated as the integral of density.
Optimization Loss
A function that quantifies the error between a model’s predictions and the target data during training.
Optimization Loss
A function that quantifies the error between a model’s predictions and the target data during training.
Overfitting
A modeling error that occurs when a function is too closely fit to a limited set of data points, resulting in poor predictive performance when applied to unseen data.
Overfitting
A modeling error that occurs when a function is too closely fit to a limited set of data points, resulting in poor predictive performance when applied to unseen data.
Neural Network
A computational model composed of interconnected layers of nodes that learns complex functions from data.
Neural Radiance Fields (NeRF)
A Radiance Field representation that reconstructs lifelike 3D scenes from 2D images by modeling the color and density of a scene.
Neural Rendering
A technique that combines deep learning models with traditional computer graphics to generate, reconstruct, or enhance images and videos.
Novel View Synthesis
The process of generating novel images of a scene from new viewpoints, based on a set of existing images.
Context: Applications include virtual reality, 3D reconstruction, and image-based rendering.
P
Participating Media
Materials like fog, smoke, or water that interact with light by absorbing and scattering it.
Participating Media
Materials like fog, smoke, or water that interact with light by absorbing and scattering it.
Path Tracing
Definition: A rendering algorithm that simulates the way light paths propagate through a scene to produce realistic images by tracing the paths of individual photons.
Context: Used in computer graphics to generate images with global illumination effects like shadows, reflections, and refractions.
Path Tracing
Definition: A rendering algorithm that simulates the way light paths propagate through a scene to produce realistic images by tracing the paths of individual photons.
Context: Used in computer graphics to generate images with global illumination effects like shadows, reflections, and refractions.
Photogrammetry
The art and science of obtaining reliable information about physical objects and the environment through the process of recording, measuring, and interpreting photographic images. Photogrammetry is not able to reconstruct reflections or highly specular scenes. Radiance Fields methods like NeRFs and Gaussian Splatting are not Photogrammetry.
Photogrammetry
The art and science of obtaining reliable information about physical objects and the environment through the process of recording, measuring, and interpreting photographic images. Photogrammetry is not able to reconstruct reflections or highly specular scenes. Radiance Fields methods like NeRFs and Gaussian Splatting are not Photogrammetry.
Photometric Consistency
The principle that corresponding points in different images should have similar brightness and color.
Photometric Consistency
The principle that corresponding points in different images should have similar brightness and color.
Photorealism
The quality of an image appearing indistinguishable from a real photograph.
Photorealism
The quality of an image appearing indistinguishable from a real photograph.
Pinhole Camera Model
An idealized model that projects 3D points onto a 2D image plane through a single point
Pinhole Camera Model
An idealized model that projects 3D points onto a 2D image plane through a single point
Plenoptic Function
Definition: A function that describes the intensity of light rays in a scene as a function of position, direction, wavelength, and time.
Context: Serves as a comprehensive representation of all visual information available in a scene.
Plenoptic Function
Definition: A function that describes the intensity of light rays in a scene as a function of position, direction, wavelength, and time.
Context: Serves as a comprehensive representation of all visual information available in a scene.
Point Cloud
A set of 3D points (often with color or normal information) representing the surface of a scene or object.
Point Cloud
A set of 3D points (often with color or normal information) representing the surface of a scene or object.
Point-Based Rendering
A technique for displaying 3D objects using individual points instead of traditional polygon meshes.
Point-Based Rendering
A technique for displaying 3D objects using individual points instead of traditional polygon meshes.
Positional Encoding
A method to enrich spatial coordinates with sinusoidal functions to capture high-frequency details.
Positional Encoding
A method to enrich spatial coordinates with sinusoidal functions to capture high-frequency details.
Pruning
In Gaussian splatting, pruning refers to the process of removing redundant or less significant Gaussian primitives to optimize memory usage and rendering efficiency. This involves evaluating the importance of each Gaussian—often based on its contribution to the overall scene—and eliminating those with minimal impact.
Pruning
In Gaussian splatting, pruning refers to the process of removing redundant or less significant Gaussian primitives to optimize memory usage and rendering efficiency. This involves evaluating the importance of each Gaussian—often based on its contribution to the overall scene—and eliminating those with minimal impact.
PSNR
An image quality metric that compares the maximum signal power to the error (noise) power expressed in decibels.
PSNR
An image quality metric that compares the maximum signal power to the error (noise) power expressed in decibels.
Neural Network
A computational model composed of interconnected layers of nodes that learns complex functions from data.
Neural Radiance Fields (NeRF)
A Radiance Field representation that reconstructs lifelike 3D scenes from 2D images by modeling the color and density of a scene.
Neural Rendering
A technique that combines deep learning models with traditional computer graphics to generate, reconstruct, or enhance images and videos.
Novel View Synthesis
The process of generating novel images of a scene from new viewpoints, based on a set of existing images.
Context: Applications include virtual reality, 3D reconstruction, and image-based rendering.
R
Radiance
Definition: The measure of the amount of light that passes through or is emitted from a particular area and falls within a given solid angle.
Context: Fundamental in radiometry, radiance quantifies the brightness of a surface as viewed from a specific direction.
Radiance
Definition: The measure of the amount of light that passes through or is emitted from a particular area and falls within a given solid angle.
Context: Fundamental in radiometry, radiance quantifies the brightness of a surface as viewed from a specific direction.
Radiance Field
A function that assigns a radiance (color intensity) value to every viewing direction in a given space.
Radiance Field
A function that assigns a radiance (color intensity) value to every viewing direction in a given space.
Radiometry
The field that measures electromagnetic radiation (including light) in absolute physical units.
Radiometry
The field that measures electromagnetic radiation (including light) in absolute physical units.
Rasterization
The process of converting vector graphics (such as triangles) into a 2D grid of pixels.
Rasterization
The process of converting vector graphics (such as triangles) into a 2D grid of pixels.
Ray Casting
A technique that sends rays into a scene to determine the first surface they hit.
Ray Casting
A technique that sends rays into a scene to determine the first surface they hit.
Ray Integration
The process of accumulating color and density values sampled along a ray to compute a pixel’s final color.
Ray Integration
The process of accumulating color and density values sampled along a ray to compute a pixel’s final color.
Ray Marching
Definition: A rendering technique used to compute the distance a ray travels through a scene before intersecting an object, often used in rendering implicit surfaces and volumetric data.
Context: Essential for visualizing complex surfaces and participating media in computer graphics.
Ray Marching
Definition: A rendering technique used to compute the distance a ray travels through a scene before intersecting an object, often used in rendering implicit surfaces and volumetric data.
Context: Essential for visualizing complex surfaces and participating media in computer graphics.
Ray Tracing
A rendering method that simulates light paths by tracing rays and their reflections/refractions to compute global illumination.
Ray Tracing
A rendering method that simulates light paths by tracing rays and their reflections/refractions to compute global illumination.
Real-time Rendering
Real time rendering is the sub-field of computer graphics focused on producing and analyzing images in real time. Typically this is thought to be a minimum of thirty frames per second.
Real-time Rendering
Real time rendering is the sub-field of computer graphics focused on producing and analyzing images in real time. Typically this is thought to be a minimum of thirty frames per second.
Refraction
The bending of light as it passes from one medium to another with a different refractive index.
Refraction
The bending of light as it passes from one medium to another with a different refractive index.
Regularization
Techniques applied during training to discourage overly complex models and reduce overfitting.
Regularization
Techniques applied during training to discourage overly complex models and reduce overfitting.
Rendering
The process of generating a 2D image from a 3D scene using computational methods.
Rendering
The process of generating a 2D image from a 3D scene using computational methods.
Rendering Equation
An integral equation that balances emitted and reflected light forming the basis of global illumination.
Rendering Equation
An integral equation that balances emitted and reflected light forming the basis of global illumination.
S
Sampling
Selecting a set of discrete points or rays from a continuous function to approximate integrals.
Sampling
Selecting a set of discrete points or rays from a continuous function to approximate integrals.
Scattering
The redirection of light as it interacts with particles or irregularities in a medium.
Scattering
The redirection of light as it interacts with particles or irregularities in a medium.
Scene Representation
The method or format used to encode the geometry and appearance of a 3D scene.
Scene Representation
The method or format used to encode the geometry and appearance of a 3D scene.
Scene Representation Netorks (SRN)
Definition: Neural networks designed to encode 3D scenes into a latent space, enabling tasks like novel view synthesis and scene manipulation.
Context: SRNs learn continuous representations of scenes, allowing for smooth interpolation between viewpoints.
Scene Representation Netorks (SRN)
Definition: Neural networks designed to encode 3D scenes into a latent space, enabling tasks like novel view synthesis and scene manipulation.
Context: SRNs learn continuous representations of scenes, allowing for smooth interpolation between viewpoints.
SDF (Signed Distance Function)
A function that returns the shortest distance from a point to a surface with the sign indicating inside or outside.
SDF (Signed Distance Function)
A function that returns the shortest distance from a point to a surface with the sign indicating inside or outside.
Shading
The calculation of surface color based on material properties, such as lighting and viewing direction.
Shading
The calculation of surface color based on material properties, such as lighting and viewing direction.
Shadow
A region of reduced light caused by an object blocking the light source.
Shadow
A region of reduced light caused by an object blocking the light source.
Signed Distance Function (SDF)
Definition: A function that returns the shortest distance from a point to a surface, with the sign indicating whether the point is inside or outside the surface.
Context: Utilized in computer graphics to represent and render complex shapes and surfaces implicitly.
Signed Distance Function (SDF)
Definition: A function that returns the shortest distance from a point to a surface, with the sign indicating whether the point is inside or outside the surface.
Context: Utilized in computer graphics to represent and render complex shapes and surfaces implicitly.
Space Carving
A reconstruction method that iteratively removes parts of a volume inconsistent with multi-view observations.
Space Carving
A reconstruction method that iteratively removes parts of a volume inconsistent with multi-view observations.
Sparse Voxel Octree
A hierarchical data structure that adaptively subdivides space into voxels storing only regions containing detail.
Sparse Voxel Octree
A hierarchical data structure that adaptively subdivides space into voxels storing only regions containing detail.
Specular Reflection
Mirror-like reflection where the angle of incidence equals the angle of reflection producing highlights.
Specular Reflection
Mirror-like reflection where the angle of incidence equals the angle of reflection producing highlights.
Spherical Harmonics
A set of orthogonal basis functions used to represent functions defined on the sphere often for approximating lighting or view dependent effects.
Spherical Harmonics
A set of orthogonal basis functions used to represent functions defined on the sphere often for approximating lighting or view dependent effects.
Structural Similarity Index Measure (SSIM)
An image quality metric that evaluates similarity based on luminance, contrast, and structural information.
Structural Similarity Index Measure (SSIM)
An image quality metric that evaluates similarity based on luminance, contrast, and structural information.
Structure from Motion (SfM)
A method for reconstructing 3D scenes from a series of 2D images taken from different viewpoints. It estimates camera positions and is often the first step in Radiance Field reconstruction.
Structure from Motion (SfM)
A method for reconstructing 3D scenes from a series of 2D images taken from different viewpoints. It estimates camera positions and is often the first step in Radiance Field reconstruction.
Subsurface Scattering
The phenomenon of light penetrating the surface of a translucent object, scattering internally, and exiting at a different point.
Subsurface Scattering
The phenomenon of light penetrating the surface of a translucent object, scattering internally, and exiting at a different point.
SuperSplat
A popular Gaussian Splatting editor, created by PlayCanvas.
SuperSplat
A popular Gaussian Splatting editor, created by PlayCanvas.
T
Temporal Consistency
The property ensuring that sequential frames (in dynamic scenes or video) change smoothly and coherently.
Temporal Consistency
The property ensuring that sequential frames (in dynamic scenes or video) change smoothly and coherently.
Tone Mapping
The process of converting high dynamic range images to a displayable range while preserving details.
Tone Mapping
The process of converting high dynamic range images to a displayable range while preserving details.
V
View Dependent Effects
Visual phenomena where an object's appearance changes based on the observer's viewpoint, such as reflections, refractions, and shading variations, giving a lifelike experience to the observer.
View Dependent Effects
Visual phenomena where an object's appearance changes based on the observer's viewpoint, such as reflections, refractions, and shading variations, giving a lifelike experience to the observer.
Volume Rendering
Definition: A method for displaying three-dimensional data without first fitting geometric primitives to the data.
Context: Commonly used in medical imaging and scientific visualization to represent complex structures.
Volume Rendering
Definition: A method for displaying three-dimensional data without first fitting geometric primitives to the data.
Context: Commonly used in medical imaging and scientific visualization to represent complex structures.
Volumetric Rendering
A technique that simulates the passage of light through a volume by integrating color and density along rays.
Volumetric Rendering
A technique that simulates the passage of light through a volume by integrating color and density along rays.
Volumetric Scene Representation
Definition: A 3D representation of a scene where space is divided into a grid (voxels), with each voxel containing information about the material properties at that location.
Context: Facilitates rendering techniques that simulate the absorption and scattering of light within a medium.
Volumetric Scene Representation
Definition: A 3D representation of a scene where space is divided into a grid (voxels), with each voxel containing information about the material properties at that location.
Context: Facilitates rendering techniques that simulate the absorption and scattering of light within a medium.