Show simple item record

dc.contributor.authorZou, Chuhang
dc.contributor.authorSu, Jheng-Wei
dc.contributor.authorPeng, Chi-Han
dc.contributor.authorColburn, Alex
dc.contributor.authorShan, Qi
dc.contributor.authorWonka, Peter
dc.contributor.authorChu, Hung-Kuo
dc.contributor.authorHoiem, Derek
dc.date.accessioned2019-12-19T07:12:08Z
dc.date.available2019-12-19T07:12:08Z
dc.date.issued2019-10-09
dc.identifier.urihttp://hdl.handle.net/10754/660690
dc.description.abstractRecent approaches for predicting layouts from 360 panoramas produce excellent results. These approaches build on a common framework consisting of three steps: a pre-processing step based on edge-based alignment, prediction of layout elements, and a post-processing step by fitting a 3D layout to the layout elements. Until now, it has been difficult to compare the methods due to multiple different design decisions, such as the encoding network (e.g. SegNet or ResNet), type of elements predicted (e.g. corners, wall/floor boundaries, or semantic segmentation), or method of fitting the 3D layout. To address this challenge, we summarize and describe the common framework, the variants, and the impact of the design decisions. For a complete evaluation, we also propose extended annotations for the Matterport3D dataset, and introduce two depth-based evaluation metrics.
dc.description.sponsorshipThis research is supported in part by ONR MURI grant N00014-16-1-2007, iStaging Corp. fund and the Ministry of Science and Technology of Taiwan (108-2218-E-007-050- and 107-2221- E-007-088-MY3). We thank Shang-Ta Yang for providing the source code of DuLa-Net. We thank Cheng Sun for providing the source code of HorizonNet and help run experiments on our provided dataset.
dc.publisherarXiv
dc.relation.urlhttps://arxiv.org/pdf/1910.04099
dc.rightsArchived with thanks to arXiv
dc.title3D Manhattan Room Layout Reconstruction from a Single 360 Image
dc.typePreprint
dc.contributor.departmentVisual Computing Center (VCC)
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.relation.referencesgithub:SunDaDenny/DuLa-Net
dc.eprint.versionPre-print
dc.contributor.institutionUniversity of Illinois at Urbana-Champaign
dc.contributor.institutionNational Tsing Hua University
dc.contributor.institutionShanghaiTech University
dc.contributor.institutionUniversity of Washington
dc.contributor.institutionApple Inc.
dc.identifier.arxivid1910.04099
kaust.personPeng, Chi-Han
kaust.personWonka, Peter
dc.relation.issupplementedbygithub:zouchuhang/LayoutNetv2
dc.relation.issupplementedbygithub:ericsujw/Matterport3DLayoutAnnotation
dc.relation.issupplementedbygithub:SunDaDenny/DuLa-Net
refterms.dateFOA2019-12-19T07:12:48Z
display.relations<b>Is Supplemented By:</b><br/> <ul><li><i>[Software]</i> <br/> Title: zouchuhang/LayoutNetv2: PyTorch implementation of our IJCV paper: "Manhattan Room Layout Reconstruction from a Single 360 image: A Comparative Study of State-of-the-art Methods". Publication Date: 2019-07-30. github: <a href="https://github.com/zouchuhang/LayoutNetv2" >zouchuhang/LayoutNetv2</a> Handle: <a href="http://hdl.handle.net/10754/667893" >10754/667893</a></a></li><li><i>[Software]</i> <br/> Title: ericsujw/Matterport3DLayoutAnnotation: Layout annotation on a subset of Matterport3D dataset. Publication Date: 2019-10-20. github: <a href="https://github.com/ericsujw/Matterport3DLayoutAnnotation" >ericsujw/Matterport3DLayoutAnnotation</a> Handle: <a href="http://hdl.handle.net/10754/667903" >10754/667903</a></a></li><b>References:</b><br/> <ul><li><i>[Software]</i> <br/> Title: SunDaDenny/DuLa-Net: Pytorch demo code of our CVPR 2019 paper: DuLa-Net: A Dual-Projection Network for Estimating Room Layouts from a Single RGB Panorama. Publication Date: 2019-10-06. github: <a href="https://github.com/SunDaDenny/DuLa-Net" >SunDaDenny/DuLa-Net</a> Handle: <a href="http://hdl.handle.net/10754/667906" >10754/667906</a></a></li></ul>


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
5.368Mb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record