Disk-NeuralRTI: Optimized NeuralRTI Relighting through Knowledge Distillation

Simple item page
dc.contributor.authorDulecha, Tinsae Gebrechristosen_US
dc.contributor.authorRighetto, Leonardoen_US
dc.contributor.authorPintus, Ruggeroen_US
dc.contributor.authorGobbetti, Enricoen_US
dc.contributor.authorGiachetti, Andreaen_US
dc.contributor.editorCaputo, Arielen_US
dc.contributor.editorGarro, Valeriaen_US
dc.contributor.editorGiachetti, Andreaen_US
dc.contributor.editorCastellani, Umbertoen_US
dc.contributor.editorDulecha, Tinsae Gebrechristosen_US
dc.date.accessioned2024-11-11T12:48:17Z
dc.date.available2024-11-11T12:48:17Z
dc.date.issued2024
dc.description.abstractRelightable images created from Multi-Light Image Collections (MLICs) are among the most employed models for interactive object exploration in cultural heritage (CH). In recent years, neural representations have been shown to produce higherquality images at similar storage costs to the more classic analytical models such as Polynomial Texture Maps (PTM) or Hemispherical Harmonics (HSH). However, the Neural RTI models proposed in the literature perform the image relighting with decoder networks with a high number of parameters, making decoding slower than for classical methods. Despite recent efforts targeting model reduction and multi-resolution adaptive rendering, exploring high-resolution images, especially on high-pixelcount displays, still requires significant resources and is only achievable through progressive rendering in typical setups. In this work, we show how, by using knowledge distillation from an original (teacher) Neural RTI network, it is possible to create a more efficient RTI decoder (student network). We evaluated the performance of the network compression approach on existing RTI relighting benchmarks, including both synthetic and real datasets, and on novel acquisitions of high-resolution images. Experimental results show that we can keep the student prediction close to the teacher with up to 80% parameter reduction and almost ten times faster rendering when embedded in an online viewer.en_US
dc.description.sectionheadersRendering
dc.description.seriesinformationSmart Tools and Applications in Graphics - Eurographics Italian Chapter Conference
dc.identifier.doi10.2312/stag.20241340
dc.identifier.isbn978-3-03868-265-3
dc.identifier.issn2617-4855
dc.identifier.pages10 pages
dc.identifier.urihttps://doi.org/10.2312/stag.20241340
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/stag20241340
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: RTI → Neural RTI, Disk-NeuralRTI; Relighting → Neural relighting; RTI Viewer → Web based visualization
dc.subjectRTI → Neural RTI
dc.subjectDisk
dc.subjectNeuralRTI
dc.subjectRelighting → Neural relighting
dc.subjectRTI Viewer → Web based visualization
dc.titleDisk-NeuralRTI: Optimized NeuralRTI Relighting through Knowledge Distillationen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
stag20241340.pdf
Size:
35.42 MB
Format:
Adobe Portable Document Format
Download
Collections
Italian Chapter Conference 2024 - Smart Tools and Apps in Graphics