Terrain Diffusion: A Diffusion-Based Successor to Perlin Noise in Infinite, Real-Time Terrain Generation

Papers with Code Trending Papers
diffusion-models procedural-generation terrain-generation infinite-generation real-time open-source

Summary

Terrain Diffusion introduces a diffusion-based successor to Perlin noise, using a novel InfiniteDiffusion algorithm to generate realistic, seamless, and boundless procedural worlds with constant-time random access.

For decades, procedural worlds have been built on procedural noise functions such as Perlin noise, which are fast and infinite, yet fundamentally limited in realism and large-scale coherence. We introduce Terrain Diffusion, an AI-era successor to Perlin noise that bridges the fidelity of diffusion models with the properties that made procedural noise indispensable: seamless infinite extent, seed-consistency, and constant-time random access. At its core is InfiniteDiffusion, a novel algorithm for infinite generation, enabling seamless, real-time synthesis of boundless landscapes. A hierarchical stack of diffusion models couples planetary context with local detail, while a compact Laplacian encoding stabilizes outputs across Earth-scale dynamic ranges. An open-source infinite-tensor framework supports constant-memory manipulation of unbounded tensors, and few-step consistency distillation enables efficient generation. Together, these components establish diffusion models as a practical foundation for procedural world generation, capable of synthesizing entire planets coherently, controllably, and without limits.
Original Article
View Cached Full Text

Cached at: 06/26/26, 05:14 PM

Paper page - Terrain Diffusion: A Diffusion-Based Successor to Perlin Noise in Infinite, Real-Time Terrain Generation

Source: https://huggingface.co/papers/2512.08309

Abstract

Terrain Diffusion uses diffusion models and a novel algorithm called InfiniteDiffusion to generate realistic, seamless, and boundless procedural worlds with constant-time random access.

For decades, procedural worlds have been built onprocedural noisefunctions such asPerlin noise, which are fast and infinite, yet fundamentally limited in realism and large-scale coherence. We introduceTerrain Diffusion, an AI-era successor toPerlin noisethat bridges the fidelity ofdiffusion modelswith the properties that madeprocedural noiseindispensable:seamless infinite extent,seed-consistency, andconstant-time random access. At its core isInfiniteDiffusion, a novel algorithm for infinite generation, enabling seamless, real-time synthesis of boundless landscapes. Ahierarchical stackofdiffusion modelscouplesplanetary contextwithlocal detail, while acompact Laplacian encodingstabilizes outputs acrossEarth-scale dynamic ranges. An open-sourceinfinite-tensor frameworksupportsconstant-memory manipulationof unbounded tensors, andfew-step consistency distillationenables efficient generation. Together, these components establishdiffusion modelsas a practical foundation for procedural world generation, capable of synthesizing entire planets coherently, controllably, and without limits.

View arXiv pageView PDFProject pageGitHub512Add to collection

Get this paper in your agent:

hf papers read 2512\.08309

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper6

#### xandergos/terrain-diffusion-30m #### xandergos/terrain-diffusion-90m #### xandergos/TerrainDiffusion-Consistency-Base-192x3 UpdatedDec 14, 2025 • 13 • 1 #### xandergos/TerrainDiffusion-Consistency-Decoder-64x3 Browse 6 models citing this paper## Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2512.08309 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2512.08309 in a Space README.md to link it from this page.

Collections including this paper2

Similar Articles

Learning to Discretize: Diffusion-Based Adaptive Mesh with Spectral Guidance

arXiv cs.LG

This paper proposes a diffusion-based framework for learning adaptive mesh discretization conditioned on observed PDE dynamics, using spectral guidance and physics constraints to allocate resolution where needed. The method achieves competitive or superior performance across five PDE regimes.

Perlin's Noise Algorithm

Hacker News Top

An explanation of Perlin's noise algorithm, covering the concept of noise, coherent noise, and the basics of implementing noise functions in procedural generation.

SynCity 3000: Bootstrapping Scene-Scale 3D Diffusion

Hugging Face Daily Papers

SynCity 3000 introduces a framework for generating large, globally coherent 3D scenes by adapting image-to-3D generators as convolutional operators, fine-tuned on synthetic scene data from a new data engine.