Displacementromp

Displacementromp is a procedural technique used in computer graphics to synthesize detailed surface displacements for 3D models. It combines displacement mapping with a sparse signal-recovery approach, using the Regularized Orthogonal Matching Pursuit (ROMP) algorithm to construct a compact, adaptable displacement field from limited data. The term arose in graphics research to describe a workflow that yields high-frequency surface detail while maintaining low memory and bandwidth requirements.

In practice, displacementromp begins with a base mesh and a set of sample displacements, either from real-world scans or procedurally generated height values. A dictionary of candidate displacement components is prepared, comprising basis functions such as localized Gaussians, wavelets, or sinusoidal taps. ROMP then identifies a small subset of these components that best approximate the target displacement while enforcing sparsity and stability. The resulting sparse field is used to drive a standard displacement map or to directly deform geometry.

Advantages include efficient storage (a few coefficients instead of dense maps), scalable detail across levels of detail, and smoother transitions between LODs. It is well-suited for real-time rendering, streaming terrains, and procedurally detailed surfaces in games and film production. Limitations involve potential artifacts in regions with rapid curvature changes if the basis is not well matched, and the need for careful tuning of the dictionary and regularization parameters. Variants may incorporate voxel-based representations or perceptual error metrics to further optimize quality.

See also: displacement mapping, sparse approximation, ROMP, real-time rendering, procedural generation.