Rendering

Rendering is the process of generating a two-dimensional image from a description of a three-dimensional scene, simulating how light interacts with surfaces and reaches a virtual camera.

Definition

Rendering is the computation of a pixel image by determining, for each pixel, the radiance arriving at the camera from the visible scene geometry under given lighting and material conditions.

Scope

This area spans the two principal families of rendering algorithms - rasterization, which projects geometry to the image plane, and ray tracing, which follows light paths through the scene - together with the physical and empirical models of light transport, surface reflectance, and shading that determine how points appear. It also covers the hardware-accelerated, real-time pipelines that make interactive graphics possible.

Sub-topics

• rasterization and the graphics pipeline
• ray tracing and global illumination
• real time and gpu rendering
• shading and reflectance models

Core questions

• Given a scene of geometry, lights, and materials, what color should each pixel be?
• How is the physics of light transport approximated efficiently enough to compute?
• What is the trade-off between physical accuracy and rendering speed?
• How can global lighting effects such as shadows, reflections, and indirect illumination be reproduced?

Key concepts

• Rasterization and ray tracing
• The rendering equation
• Radiance and irradiance
• Local and global illumination
• Visibility and hidden-surface removal
• Shading and reflectance

Key theories

The rendering equation

Light transport in a scene is governed by an integral equation expressing outgoing radiance at a point as the sum of emitted radiance and reflected incoming radiance integrated over the hemisphere, providing the unifying physical foundation for photorealistic rendering.

Local versus global illumination

Local illumination shades each surface point using only direct light sources, while global illumination additionally accounts for light that bounces between surfaces, producing soft shadows, color bleeding, and caustics at substantially higher computational cost.

Clinical relevance

Rendering underpins film visual effects and animation, video games, architectural and product visualization, virtual and augmented reality, scientific and medical visualization, and the synthetic-data pipelines used to train computer-vision systems.

History

Early raster graphics in the 1970s established hidden-surface and shading algorithms; Whitted's 1980 recursive ray tracing and Kajiya's 1986 rendering equation formalized light transport, and the rise of programmable GPUs from the 2000s onward brought physically based and real-time rendering into the mainstream.

Debates

Rasterization versus ray tracing for real-time graphics

Rasterization has long dominated interactive rendering for its speed, while ray tracing offers more naturally correct shadows, reflections, and global illumination; hardware ray-tracing acceleration has narrowed but not closed the performance gap, leaving hybrid pipelines common.

Key figures

• James Kajiya
• Turner Whitted
• Edwin Catmull

Seminal works

• kajiya1986
• pharr2016
• hughes2013

Frequently asked questions

What is the difference between rasterization and ray tracing?

Rasterization projects scene geometry onto the image and fills in pixels it covers, which is fast; ray tracing instead follows rays from the camera into the scene to find what each pixel sees, which more naturally captures reflections, refractions, and shadows but costs more.

Why do photorealistic renders take so long?

Simulating light that bounces many times between surfaces requires evaluating high-dimensional integrals, usually by tracing millions of randomized light paths and averaging them, so reducing noise to an acceptable level is computationally expensive.