Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Improved Stochastic Texture Filtering Through Sample Reuse

Wronski, Bartlomiej ; Pharr, Matt and Akenine-Möller, Tomas LU (2025) In Proceedings of the ACM on Computer Graphics and Interactive Techniques 8(1).
Abstract

Stochastic texture filtering (STF) has re-emerged as a technique that can bring down the cost of texture filtering of advanced texture compression methods, e.g., neural texture compression. However, during texture magnification, the swapped order of filtering and shading with STF can result in aliasing. The inability to smoothly interpolate material properties stored in textures, such as surface normals, leads to potentially undesirable appearance changes. We present a novel method to improve the quality of stochastically-filtered magnified textures and reduce the image difference compared to traditional texture filtering. When textures are magnified, nearby pixels filter similar sets of texels and we introduce techniques for sharing... (More)

Stochastic texture filtering (STF) has re-emerged as a technique that can bring down the cost of texture filtering of advanced texture compression methods, e.g., neural texture compression. However, during texture magnification, the swapped order of filtering and shading with STF can result in aliasing. The inability to smoothly interpolate material properties stored in textures, such as surface normals, leads to potentially undesirable appearance changes. We present a novel method to improve the quality of stochastically-filtered magnified textures and reduce the image difference compared to traditional texture filtering. When textures are magnified, nearby pixels filter similar sets of texels and we introduce techniques for sharing texel values among pixels with only a small increase in cost (0.04-0.14 ms per frame). We propose an improvement to weighted importance sampling that guarantees that our method never increases error beyond single-sample stochastic texture filtering. Under high magnification, our method has >10 dB higher PSNR than single-sample STF. Our results show greatly improved image quality both with and without spatiotemporal denoising.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
importance sampling, stochastic texture filtering, wave intrinsics
in
Proceedings of the ACM on Computer Graphics and Interactive Techniques
volume
8
issue
1
article number
14
publisher
Association for Computing Machinery (ACM)
external identifiers
  • scopus:105006613709
ISSN
2577-6193
DOI
10.1145/3728292
language
English
LU publication?
yes
additional info
Publisher Copyright: © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.
id
eed34fb7-fe66-49d5-8d9b-6d9ebc041923
date added to LUP
2025-12-18 13:34:29
date last changed
2025-12-18 13:35:44
@article{eed34fb7-fe66-49d5-8d9b-6d9ebc041923,
  abstract     = {{<p>Stochastic texture filtering (STF) has re-emerged as a technique that can bring down the cost of texture filtering of advanced texture compression methods, e.g., neural texture compression. However, during texture magnification, the swapped order of filtering and shading with STF can result in aliasing. The inability to smoothly interpolate material properties stored in textures, such as surface normals, leads to potentially undesirable appearance changes. We present a novel method to improve the quality of stochastically-filtered magnified textures and reduce the image difference compared to traditional texture filtering. When textures are magnified, nearby pixels filter similar sets of texels and we introduce techniques for sharing texel values among pixels with only a small increase in cost (0.04-0.14 ms per frame). We propose an improvement to weighted importance sampling that guarantees that our method never increases error beyond single-sample stochastic texture filtering. Under high magnification, our method has &gt;10 dB higher PSNR than single-sample STF. Our results show greatly improved image quality both with and without spatiotemporal denoising.</p>}},
  author       = {{Wronski, Bartlomiej and Pharr, Matt and Akenine-Möller, Tomas}},
  issn         = {{2577-6193}},
  keywords     = {{importance sampling; stochastic texture filtering; wave intrinsics}},
  language     = {{eng}},
  month        = {{05}},
  number       = {{1}},
  publisher    = {{Association for Computing Machinery (ACM)}},
  series       = {{Proceedings of the ACM on Computer Graphics and Interactive Techniques}},
  title        = {{Improved Stochastic Texture Filtering Through Sample Reuse}},
  url          = {{http://dx.doi.org/10.1145/3728292}},
  doi          = {{10.1145/3728292}},
  volume       = {{8}},
  year         = {{2025}},
}