Edge-preserving noise for diffusion models

Simple item page
dc.contributor.authorVandersanden, Jente
dc.contributor.authorHoll, Sascha
dc.contributor.authorHuang, Xingchang
dc.contributor.authorSingh, Gurprit
dc.contributor.editorMasia, Belen
dc.contributor.editorThies, Justus
dc.date.accessioned2026-04-17T13:38:24Z
dc.date.available2026-04-17T13:38:24Z
dc.date.issued2026
dc.description.abstractClassical diffusion models typically rely on isotropic Gaussian noise, treating all regions uniformly and overlooking structural information important for high-quality generation. We introduce an edge-preserving diffusion process that generalizes isotropic models via a hybrid noise scheme with an edge-aware scheduler that smoothly transitions from edge-preserving to isotropic noise. This enables the model to capture fine structural details while generally maintaining global performance. We evaluate the impact of structure-aware noise in both diffusion and flow-matching frameworks, and show that existing isotropic models can be efficiently fine-tuned with edge-preserving noise, making our framework practical for adapting pre-trained systems. Beyond unconditional generation, our method particularly shows improvements in structure-guided tasks such as stroke-to-image synthesis, improving robustness and perceptual quality, as evidenced by consistent improvements across FID, KID, and CLIP-score.
dc.description.number2
dc.description.sectionheadersDiffusion and Beyond: Controlled Image Generation and Stylization
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume45
dc.identifier.doi10.1111/cgf.70383
dc.identifier.issn1467-8659
dc.identifier.pages13 pages
dc.identifier.urihttps://diglib.eg.org/handle/10.1111/cgf70383
dc.identifier.urihttps://doi.org/10.1111/cgf.70383
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.
dc.rightsCC-BY-4.0
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectGaussian processes
dc.subjectImage processing
dc.subjectMarkov processes
dc.titleEdge-preserving noise for diffusion models
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
cgf70383.pdf
Size:
37.63 MB
Format:
Adobe Portable Document Format
Download
Collections
45-Issue 2
EG 2026 - Full Papers - CGF 45-Issue 2