Saved in:
Bibliographic Details
Main Authors: Maesumi, Arman, Hu, Dylan, Saripalli, Krishi, Kim, Vladimir G., Fisher, Matthew, Pirk, Sören, Ritchie, Daniel
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.16292
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911852835897344
author Maesumi, Arman
Hu, Dylan
Saripalli, Krishi
Kim, Vladimir G.
Fisher, Matthew
Pirk, Sören
Ritchie, Daniel
author_facet Maesumi, Arman
Hu, Dylan
Saripalli, Krishi
Kim, Vladimir G.
Fisher, Matthew
Pirk, Sören
Ritchie, Daniel
contents Procedural noise is a fundamental component of computer graphics pipelines, offering a flexible way to generate textures that exhibit "natural" random variation. Many different types of noise exist, each produced by a separate algorithm. In this paper, we present a single generative model which can learn to generate multiple types of noise as well as blend between them. In addition, it is capable of producing spatially-varying noise blends despite not having access to such data for training. These features are enabled by training a denoising diffusion model using a novel combination of data augmentation and network conditioning techniques. Like procedural noise generators, the model's behavior is controllable via interpretable parameters and a source of randomness. We use our model to produce a variety of visually compelling noise textures. We also present an application of our model to improving inverse procedural material design; using our model in place of fixed-type noise nodes in a procedural material graph results in higher-fidelity material reconstructions without needing to know the type of noise in advance.
format Preprint
id arxiv_https___arxiv_org_abs_2404_16292
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle One Noise to Rule Them All: Learning a Unified Model of Spatially-Varying Noise Patterns
Maesumi, Arman
Hu, Dylan
Saripalli, Krishi
Kim, Vladimir G.
Fisher, Matthew
Pirk, Sören
Ritchie, Daniel
Graphics
Computer Vision and Pattern Recognition
Machine Learning
Procedural noise is a fundamental component of computer graphics pipelines, offering a flexible way to generate textures that exhibit "natural" random variation. Many different types of noise exist, each produced by a separate algorithm. In this paper, we present a single generative model which can learn to generate multiple types of noise as well as blend between them. In addition, it is capable of producing spatially-varying noise blends despite not having access to such data for training. These features are enabled by training a denoising diffusion model using a novel combination of data augmentation and network conditioning techniques. Like procedural noise generators, the model's behavior is controllable via interpretable parameters and a source of randomness. We use our model to produce a variety of visually compelling noise textures. We also present an application of our model to improving inverse procedural material design; using our model in place of fixed-type noise nodes in a procedural material graph results in higher-fidelity material reconstructions without needing to know the type of noise in advance.
title One Noise to Rule Them All: Learning a Unified Model of Spatially-Varying Noise Patterns
topic Graphics
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2404.16292