UTOPIC: Uncertainty-aware Overlap Prediction Network for Partial Point Cloud Registration

Simple item page
dc.contributor.authorChen, Zhileien_US
dc.contributor.authorChen, Honghuaen_US
dc.contributor.authorGong, Linaen_US
dc.contributor.authorYan, Xuefengen_US
dc.contributor.authorWang, Junen_US
dc.contributor.authorGuo, Yanwenen_US
dc.contributor.authorQin, Jingen_US
dc.contributor.authorWei, Mingqiangen_US
dc.contributor.editorUmetani, Nobuyukien_US
dc.contributor.editorWojtan, Chrisen_US
dc.contributor.editorVouga, Etienneen_US
dc.date.accessioned2022-10-04T06:39:32Z
dc.date.available2022-10-04T06:39:32Z
dc.date.issued2022
dc.description.abstractHigh-confidence overlap prediction and accurate correspondences are critical for cutting-edge models to align paired point clouds in a partial-to-partial manner. However, there inherently exists uncertainty between the overlapping and non-overlapping regions, which has always been neglected and significantly affects the registration performance. Beyond the current wisdom, we propose a novel uncertainty-aware overlap prediction network, dubbed UTOPIC, to tackle the ambiguous overlap prediction problem; to our knowledge, this is the first to explicitly introduce overlap uncertainty to point cloud registration. Moreover, we induce the feature extractor to implicitly perceive the shape knowledge through a completion decoder, and present a geometric relation embedding for Transformer to obtain transformation-invariant geometry-aware feature representations.With the merits of more reliable overlap scores and more precise dense correspondences, UTOPIC can achieve stable and accurate registration results, even for the inputs with limited overlapping areas. Extensive quantitative and qualitative experiments on synthetic and real benchmarks demonstrate the superiority of our approach over state-of-the-art methods.en_US
dc.description.number7
dc.description.sectionheadersPoint Cloud Processing and Dataset Generation
dc.description.seriesinformationComputer Graphics Forum
dc.description.volume41
dc.identifier.doi10.1111/cgf.14659
dc.identifier.issn1467-8659
dc.identifier.pages87-98
dc.identifier.pages12 pages
dc.identifier.urihttps://doi.org/10.1111/cgf.14659
dc.identifier.urihttps://diglib.eg.org:443/handle/10.1111/cgf14659
dc.publisherThe Eurographics Association and John Wiley & Sons Ltd.en_US
dc.subjectCCS Concepts: Computing methodologies → Point-based models
dc.subjectComputing methodologies → Point
dc.subjectbased models
dc.titleUTOPIC: Uncertainty-aware Overlap Prediction Network for Partial Point Cloud Registrationen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
v41i7pp087-098.pdf
Size:
7.59 MB
Format:
Adobe Portable Document Format
Download
Collections
41-Issue 7