Show simple item record

dc.contributor.authorZhang, Biao
dc.contributor.authorWonka, Peter
dc.date.accessioned2019-12-22T13:56:12Z
dc.date.available2019-12-22T13:56:12Z
dc.date.issued2019-11-30
dc.identifier.urihttp://hdl.handle.net/10754/660736
dc.description.abstractIn this paper we propose a new framework for point cloud instance segmentation. Our framework has two steps: an embedding step and a clustering step. In the embedding step, our main contribution is to propose a probabilistic embedding space for point cloud embedding. Specifically, each point is represented as a tri-variate normal distribution. In the clustering step, we propose a novel loss function, which benefits both the semantic segmentation and the clustering. Our experimental results show important improvements to the SOTA, i.e., 3.1% increased average per-category mAP on the PartNet dataset.
dc.publisherarXiv
dc.relation.urlhttps://arxiv.org/pdf/1912.00145
dc.rightsArchived with thanks to arXiv
dc.titlePoint Cloud Instance Segmentation using Probabilistic Embeddings
dc.typePreprint
dc.contributor.departmentKing Abdullah University for Science and Technology
dc.contributor.departmentComputer Science Program
dc.contributor.departmentVisual Computing Center (VCC)
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.eprint.versionPre-print
dc.identifier.arxivid1912.00145
kaust.personZhang, Biao
kaust.personWonka, Peter
refterms.dateFOA2019-12-22T13:58:11Z


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
28.80Mb
Format:
PDF
Description:
Pre-print
Thumbnail
Name:
Preprintfile1.pdf
Size:
28.80Mb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record