PlenopticPoints: Rasterizing Neural Feature Points for High-Quality Novel View Synthesis

Simple item page
dc.contributor.authorHahlbohm, Florianen_US
dc.contributor.authorKappel, Moritzen_US
dc.contributor.authorTauscher, Jan-Philippen_US
dc.contributor.authorEisemann, Martinen_US
dc.contributor.authorMagnor, Marcusen_US
dc.contributor.editorGuthe, Michaelen_US
dc.contributor.editorGrosch, Thorstenen_US
dc.date.accessioned2023-09-25T11:36:53Z
dc.date.available2023-09-25T11:36:53Z
dc.date.issued2023
dc.description.abstractThis paper presents a point-based, neural rendering approach for complex real-world objects from a set of photographs. Our method is specifically geared towards representing fine detail and reflective surface characteristics at improved quality over current state-of-the-art methods. From the photographs, we create a 3D point model based on optimized neural feature points located on a regular grid. For rendering, we employ view-dependent spherical harmonics shading, differentiable rasterization, and a deep neural rendering network. By combining a point-based approach and novel regularizers, our method is able to accurately represent local detail such as fine geometry and high-frequency texture while at the same time convincingly interpolating unseen viewpoints during inference. Our method achieves about 7 frames per second at 800×800 pixel output resolution on commodity hardware, putting it within reach for real-time rendering applications.en_US
dc.description.sectionheadersRendering and Modelling
dc.description.seriesinformationVision, Modeling, and Visualization
dc.identifier.doi10.2312/vmv.20231226
dc.identifier.isbn978-3-03868-232-5
dc.identifier.pages53-61
dc.identifier.pages9 pages
dc.identifier.urihttps://doi.org/10.2312/vmv.20231226
dc.identifier.urihttps://diglib.eg.org:443/handle/10.2312/vmv20231226
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies → Image-based rendering; Point-based models
dc.subjectComputing methodologies → Image
dc.subjectbased rendering
dc.subjectPoint
dc.subjectbased models
dc.titlePlenopticPoints: Rasterizing Neural Feature Points for High-Quality Novel View Synthesisen_US
Files
Original bundle
Now showing 1 - 3 of 3
Thumbnail Image
Name:
053-061.pdf
Size:
20.66 MB
Format:
Adobe Portable Document Format
Download
No Thumbnail Available
Name:
paper1021_mm.mp4
Size:
52.84 MB
Format:
Unknown data format
Download
Thumbnail Image
Name:
paper1021_mm.pdf
Size:
6.9 MB
Format:
Adobe Portable Document Format
Download
Collections
VMV2023