Glossary

The 3D asset pipeline is the formalized sequence of tools, processes, and data transformations that move a 3D asset from its initial creation to final delivery to viewers.

3D PBR (physically based rendering) is a rendering approach that simulates how light interacts with real-world materials.

A 3D product configurator lets users customize and visualize products interactively — changing colors, materials, and options live in the browser.

3D streaming delivers three-dimensional content progressively over a network, so you can start interacting immediately while quality improves over time.

Adaptive streaming monitors network conditions in real time and selects the best content quality the viewer's connection can reliably handle.

Anisotropic reflection is reflection whose intensity and shape vary with the direction of a surface's microscopic structure, rather than looking the same when the surface is rotated in place.

Asset optimization prepares 3D models for real-time delivery via geometry simplification, compression, and LOD generation for web and mobile.

Augmented reality overlays digital content onto the real world in real time, keeping physical environments visible while adding 3D objects and spatial context.

A function describing how a surface reflects light from one direction toward a viewer in another — the core of physically based materials.

Diffuse reflection is the scattering of incoming light in many directions when it hits a rough or matte surface, so the surface looks similar from most viewing angles.

A virtual, interactive 3D replica of a real-world object, space, or system, kept in sync with its physical counterpart.

An open-source library from Google for compressing 3D meshes and point clouds so they transfer and load faster.

Gaussian splatting models 3D scenes as collections of oriented Gaussian functions, enabling fast, photorealistic rendering from any viewpoint.

GL Transmission Format (glTF) is an open standard for 3D models, optimized for efficient real-time transmission in web and interactive apps.

A container format for GPU-ready, supercompressed textures that load directly to the GPU without an expensive decode step.

Lambertian reflectance describes a surface that scatters incoming light evenly in all directions, so it appears equally bright from every viewing angle.

LOD adjusts 3D model complexity by viewing distance — showing high detail up close and simplified geometry farther away to balance quality and performance.

A sensing technology that measures distance with laser pulses to produce a precise 3D point cloud of a scene.

MaterialX is an open standard for authoring and exchanging material and shader graphs across 3D applications and renderers in a renderer-independent, node-graph format.

Neural radiance fields (NeRFs) represent 3D scenes as neural networks trained on photographs, enabling novel-view synthesis without polygon geometry.

A non-Lambertian surface is a material whose appearance changes with viewing angle, because it reflects or refracts light instead of scattering it evenly in all directions.

Generating new, photorealistic views of a scene from camera angles that were never actually photographed.

OpenEXR is ILM's open-source high dynamic range image format for professional VFX, rendering, and compositing, with floating-point precision, multi-channel AOVs, and deep image support.

OpenPBR is an open, standardized shading model for physically based rendering, enabling consistent material appearance and interchange across 3D tools, renderers, and real-time engines.

OpenUSD is an open-source 3D scene description framework for interchanging, composing, and simulating complex 3D scenes across different software tools.

A capture technique that reconstructs a 3D model from overlapping photographs taken from multiple angles.

Photoreal rendering produces 3D images indistinguishable from photographs by accurately simulating light, shadows, materials, and surface behavior.

Pixel streaming renders 3D graphics on remote servers and streams the output as compressed video. Learn how it works and how it compares to 3D streaming.

A 3D representation made of many points in space, each with a position and often color, describing an object or scene.

A function describing how light travels in every direction through every point in a 3D space, enabling rendering from any viewpoint.

The process of converting 3D geometry into the 2D pixels shown on screen — the dominant real-time rendering technique.

Refraction is the bending of light as it passes from one transparent medium into another, such as from air into glass or water.

Spatial computing places digital objects in real-world space, merging physical and digital environments through AR, VR, and real-time environment tracking.

Specular reflection is the mirror-like bouncing of light off a smooth surface in a single dominant direction, governed by the angle of incidence.

Mathematical functions that compactly encode how color or lighting varies across all directions — used for view-dependent color in Gaussian splatting.

A 3D model processed for progressive delivery over a network, so it loads and renders in a browser or app without a full download first.

A single-file, zipped package of OpenUSD content built for easy sharing and AR viewing, especially on Apple devices.

Video that captures a moving scene in 3D over time, so it can be viewed from any angle rather than from one fixed camera.

A web 3D model viewer renders 3D asset files interactively in a browser — via a JavaScript library, embeddable widget, or web app.

WebGL is a JavaScript API that enables interactive 2D and 3D graphics in web browsers via GPU acceleration — no plugins or installs required.

A modern web graphics API that gives browsers lower-level, higher-performance GPU access than WebGL, built for real-time, high-fidelity 3D.

A browser API for building virtual and augmented reality experiences that run on the web, without installing a native app.