Show simple item record

dc.contributor.authorQian, Guocheng
dc.contributor.authorAbualshour, Abdulellah
dc.contributor.authorLi, Guohao
dc.contributor.authorThabet, Ali Kassem
dc.contributor.authorGhanem, Bernard
dc.date.accessioned2021-11-03T12:35:33Z
dc.date.available2019-12-22T11:50:32Z
dc.date.available2021-11-03T12:35:33Z
dc.date.issued2021-11-02
dc.identifier.citationQian, G., Abualshour, A., Li, G., Thabet, A., & Ghanem, B. (2021). PU-GCN: Point Cloud Upsampling using Graph Convolutional Networks. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). doi:10.1109/cvpr46437.2021.01151
dc.identifier.isbn978-1-6654-4510-8
dc.identifier.issn1063-6919
dc.identifier.doi10.1109/CVPR46437.2021.01151
dc.identifier.urihttp://hdl.handle.net/10754/660725
dc.description.abstractThe effectiveness of learning-based point cloud upsampling pipelines heavily relies on the upsampling modules and feature extractors used therein. For the point upsampling module, we propose a novel model called NodeShuffle, which uses a Graph Convolutional Network (GCN) to better encode local point information from point neighborhoods. NodeShuffle is versatile and can be incorporated into any point cloud upsampling pipeline. Extensive experiments show how NodeShuffle consistently improves state-of-the-art upsampling methods. For feature extraction, we also propose a new multi-scale point feature extractor, called Inception DenseGCN. By aggregating features at multiple scales, this feature extractor enables further performance gain in the final upsampled point clouds. We combine Inception DenseGCN with NodeShuffle into a new point upsampling pipeline called PU-GCN. PU-GCN sets new state-of-art performance with much fewer parameters and more efficient inference. Our code is publicly available at https://github.com/guochengqian/PU-GCN.
dc.description.sponsorshipThe authors thank Silvio Giancola for his help with the project. This work was supported by the King Abdullah University of Science and Technology (KAUST) Office of Sponsored Research through the Visual Computing Center (VCC) funding.
dc.publisherIEEE
dc.relation.urlhttps://ieeexplore.ieee.org/document/9577901/
dc.relation.urlhttps://ieeexplore.ieee.org/document/9577901/
dc.relation.urlhttps://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=9577901
dc.rightsArchived with thanks to IEEE
dc.titlePU-GCN: Point Cloud Upsampling using Graph Convolutional Networks
dc.typeConference Paper
dc.contributor.departmentComputer Science
dc.contributor.departmentComputer, Electrical and Mathematical Science and Engineering (CEMSE) Division
dc.contributor.departmentKing Abdullah University of Science and Technology (KAUST)
dc.contributor.departmentVisual Computing Center (VCC)
dc.contributor.departmentElectrical and Computer Engineering Program
dc.conference.date20-25 June 2021
dc.conference.name2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
dc.conference.locationNashville, TN, USA
dc.eprint.versionPost-print
dc.identifier.arxivid1912.03264
kaust.personQian, Guocheng
kaust.personAbualshour, Abdulellah
kaust.personLi, Guohao
kaust.personThabet, Ali Kassem
kaust.personGhanem, Bernard
dc.relation.issupplementedbygithub:guochengqian/PU-GCN
refterms.dateFOA2019-12-22T11:51:22Z
display.relations<b>Is Supplemented By:</b><br/> <ul><li><i>[Software]</i> <br/> Title: guochengqian/PU-GCN: The official implementation of PU-GCN https://sites.google.com/kaust.edu.sa/pugcn. Publication Date: 2019-11-19. github: <a href="https://github.com/guochengqian/PU-GCN" >guochengqian/PU-GCN</a> Handle: <a href="http://hdl.handle.net/10754/668102" >10754/668102</a></a></li></ul>
kaust.acknowledged.supportUnitOffice of Sponsored Research
kaust.acknowledged.supportUnitVisual Computing Center (VCC)
dc.date.published-online2021-11-02
dc.date.published-print2021-06
dc.date.posted2019-11-30


Files in this item

Thumbnail
Name:
CVPR21_PU_GCN.pdf
Size:
6.070Mb
Format:
PDF
Description:
Accepted manuscript

This item appears in the following Collection(s)

Show simple item record

VersionItemEditorDateSummary

*Selected version