Show simple item record

dc.contributor.authorLu, Yuzheen_US
dc.contributor.authorJiang, Kairongen_US
dc.contributor.authorLevine, Joshua A.en_US
dc.contributor.authorBerger, Matthewen_US
dc.contributor.editorBorgo, Rita and Marai, G. Elisabeta and Landesberger, Tatiana vonen_US
dc.date.accessioned2021-06-12T11:01:29Z
dc.date.available2021-06-12T11:01:29Z
dc.date.issued2021
dc.identifier.issn1467-8659
dc.identifier.urihttps://doi.org/10.1111/cgf.14295
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf14295
dc.description.abstractWe present an approach for compressing volumetric scalar fields using implicit neural representations. Our approach represents a scalar field as a learned function, wherein a neural network maps a point in the domain to an output scalar value. By setting the number of weights of the neural network to be smaller than the input size, we achieve compressed representations of scalar fields, thus framing compression as a type of function approximation. Combined with carefully quantizing network weights, we show that this approach yields highly compact representations that outperform state-of-the-art volume compression approaches. The conceptual simplicity of our approach enables a number of benefits, such as support for time-varying scalar fields, optimizing to preserve spatial gradients, and random-access field evaluation. We study the impact of network design choices on compression performance, highlighting how simple network architectures are effective for a broad range of volumes.en_US
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectHuman centered computing
dc.subjectVisualization
dc.subjectComputing methodologies
dc.subjectNeural networks
dc.subjectImage compression
dc.titleCompressive Neural Representations of Volumetric Scalar Fieldsen_US
dc.description.seriesinformationComputer Graphics Forum
dc.description.sectionheadersVolume and Vector Computing and Representation
dc.description.volume40
dc.description.number3
dc.identifier.doi10.1111/cgf.14295
dc.identifier.pages135-146


Files in this item

Thumbnail

This item appears in the following Collection(s)

  • 40-Issue 3
    EuroVis 2021 - Conference Proceedings

Show simple item record