What Are Computer Graphics?

Computer graphics is a field of computer science concerned with the synthesis, manipulation, and display of visual information using computational methods. It covers the full pipeline from mathematical representation of geometric objects, through the simulation of lighting and material properties, to the production of images on display devices. The field is foundational to interactive software, scientific visualization, film and broadcast production, and industrial design. Its methods rely on linear algebra, computational geometry, numerical analysis, and signal processing, and its outputs are rendered through graphics processing units (GPUs) that execute parallel shading programs at high throughput.

The discipline emerged as a formal research area in the early 1960s, with Ivan Sutherland's Sketchpad system (1963) establishing interactive graphical editing as a viable paradigm. IEEE supports the field through the IEEE Transactions on Visualization and Computer Graphics and through the IEEE Computer Society's technical committee on computer graphics, reflecting the breadth of the field across hardware, algorithms, and applications.

Rendering and Ray Tracing

Rendering is the process of computing a two-dimensional image from a description of a three-dimensional scene, including geometry, material reflectance properties, light sources, and camera parameters. Rasterization dominates real-time applications: the scene's triangular geometry is projected onto the image plane, and programmable GPU shader programs compute the color of each resulting pixel using lighting models. Ray tracing simulates light transport more faithfully by tracing paths from the camera through each pixel into the scene, computing intersections with geometry and evaluating reflections, refractions, and shadows physically. Ray tracing was historically too slow for real-time use, but dedicated hardware units in modern GPUs and techniques such as denoising have made hybrid rasterization-plus-raytracing pipelines practical in interactive applications. Path tracing, a Monte Carlo extension of ray tracing, estimates the full global illumination integral by sampling many light paths per pixel, producing photorealistic results used in film visual effects and architectural visualization. An IEEE Xplore paper on emerging trends in the graphics rendering pipeline surveys how hardware and algorithm co-design has advanced real-time rendering quality.

Geometric Modeling and Mesh Generation

Before rendering can occur, the shapes of objects in a scene must be represented in a form the computer can process. Polygon meshes, collections of vertices connected by edges and faces, are the dominant surface representation because GPU hardware is optimized for triangle rasterization. Smooth surfaces are approximated using curve and surface fitting methods: spline curves and NURBS (non-uniform rational B-splines) define smooth paths and surfaces from a compact set of control points, and subdivision surfaces iteratively refine a coarse mesh to approach a smooth limit. Mesh generation from volumetric data or sampled point clouds uses algorithms from computational geometry, such as Delaunay triangulation, which maximizes the minimum angle of triangles and reduces numerical conditioning problems in downstream simulation. Solid modeling systems such as those based on constructive solid geometry (CSG) represent objects as Boolean combinations of primitive shapes, enabling precise dimensional control needed in engineering design.

Character Generation and Animation

Animating characters requires representing deformable geometry and driving it with physically plausible or artistically controlled motion. Skeletal animation binds a mesh to a hierarchy of bones; animators keyframe the rotations of bones, and a skinning algorithm interpolates the mesh vertices attached to each bone to produce the deformed shape. Motion capture systems record the positions of markers attached to a human performer, providing high-fidelity reference motion that is retargeted to the proportions of a digital character. Procedural animation generates motion from physical simulation or rule systems, useful for secondary effects such as cloth, hair dynamics, and crowd behavior where keyframing thousands of independently moving elements is impractical. ACM SIGGRAPH proceedings document the research contributions that have shaped character animation methods over the past four decades.

Applications

Computer graphics has applications in a wide range of fields, including:

• Film and broadcast visual effects, where photorealistic digital imagery is composited with live footage
• Video games, where real-time rendering must sustain 60 or more frames per second
• Scientific and engineering visualization, where simulation results are rendered on high-resolution workstations
• Architectural design and urban planning, where models are visualized before construction
• Digital art and interactive installations, where generative visual systems produce dynamic imagery