MDNF: Multi-Diffusion-Nets for Neural Fields on Meshes

dc.contributor.authorRimon, Avigail Cohenen_US
dc.contributor.authorShnitzer, Talen_US
dc.contributor.authorBen-Chen, Mirelaen_US
dc.contributor.editorAttene, Marcoen_US
dc.contributor.editorSellán, Silviaen_US
dc.date.accessioned2025-06-20T07:39:25Z
dc.date.available2025-06-20T07:39:25Z
dc.date.issued2025
dc.description.abstractWe propose a novel framework for representing neural fields on triangle meshes that is multi-resolution across both spatial and frequency domains. Inspired by the Neural Fourier Filter Bank (NFFB), our architecture decomposes the spatial and frequency domains by associating finer spatial resolution levels with higher frequency bands, while coarser resolutions are mapped to lower frequencies. To achieve geometry-aware spatial decomposition we leverage multiple DiffusionNet components, each associated with a different spatial resolution level. Subsequently, we apply a Fourier feature mapping to encourage finer resolution levels to be associated with higher frequencies. The final signal is composed in a wavelet-inspired manner using a sine-activated MLP, aggregating higher-frequency signals on top of lower-frequency ones. Our architecture attains high accuracy in learning complex neural fields and is robust to discontinuities, exponential scale variations of the target field, and mesh modification. We demonstrate the effectiveness of our approach through its application to diverse neural fields, such as synthetic RGB functions, UV texture coordinates, and vertex normals, illustrating different challenges. To validate our method, we compare its performance against two alternatives, showcasing the advantages of our multi-resolution architecture.en_US
dc.description.number5
dc.description.sectionheadersFields on Meshes
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume44
dc.identifier.doi10.1111/cgf.70188
dc.identifier.issn1467-8659
dc.identifier.pages14 pages
dc.identifier.urihttps://doi.org/10.1111/cgf.70188
dc.identifier.urihttps://diglib.eg.org/handle/10.1111/cgf70188
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies → Machine Learning, Shape Analysis
dc.subjectComputing methodologies → Machine Learning
dc.subjectShape Analysis
dc.titleMDNF: Multi-Diffusion-Nets for Neural Fields on Meshesen_US
Files
Original bundle
Now showing 1 - 2 of 2
Loading...
Thumbnail Image
Name:
cgf70188.pdf
Size:
74.63 MB
Format:
Adobe Portable Document Format
Loading...
Thumbnail Image
Name:
fp1-1029_mm.pdf
Size:
12.07 MB
Format:
Adobe Portable Document Format
Collections