360ViewPET: View Based Pose EsTimation for Ultra-Sparse 360-Degree Cameras

Qian Zhou; Bo Chen; Zhe Yang; Hongpeng Guo; Klara Nahrstedt

doi:10.1109/ISM52913.2021.00008

360ViewPET: View Based Pose EsTimation for Ultra-Sparse 360-Degree Cameras

Qian Zhou, Bo Chen, Zhe Yang, Hongpeng Guo, Klara Nahrstedt

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

1 Citation (Scopus)

Abstract

Immersive virtual tours based on 360-degree cameras, showing famous outdoor scenery, are becoming more and more desirable due to travel costs, pandemics and other constraints. To feel immersive, a user must receive the view accurately corresponding to her position and orientation in the virtual space when she moves inside, and this requires cameras' orientations to be known. Outdoor tour contexts have numerous, ultra-sparse cameras deployed across a wide area, making camera pose estimation challenging. As a result, pose estimation techniques like SLAM, which require mobile or dense cameras, are not applicable. In this paper we present a novel strategy called 360ViewPET, which automatically estimates the relative poses of two stationary, ultra-sparse (15 meters apart) 360-degree cameras using one equirectangular image taken by each camera. Our experiments show that it achieves accurate pose estimation, with a mean error as low as 0.9 degree. © 2021 IEEE.

Original language	English
Title of host publication	Proceedings - 23rd IEEE International Symposium on Multimedia, ISM 2021
Place of Publication	Los Alamitos, Calif.
Publisher	IEEE
Pages	1-8
ISBN (Electronic)	9781665437349
ISBN (Print)	9781665437356
DOIs	https://doi.org/10.1109/ISM52913.2021.00008
Publication status	Published - Nov 2021
Externally published	Yes
Event	23rd IEEE International Symposium on Multimedia (ISM 2021) - Virtual Duration: 29 Nov 2021 → 1 Dec 2021 https://semanticcomputing.wixsite.com/website-15

Publication series

Name	Proceedings - IEEE International Symposium on Multimedia, ISM

Conference

Conference	23rd IEEE International Symposium on Multimedia (ISM 2021)
Abbreviated title	ISM2021
Period	29/11/21 → 1/12/21
Internet address	https://semanticcomputing.wixsite.com/website-15

Funding

This research was funded by the National Science Foundation CNS 1900875, by the Postdoctoral Fellowship Program at CS UIUC and by the Grainger College of Engineering funding.

Research Keywords

360 camera
pose estimation
virtual tourism

Access Output

10.1109/ISM52913.2021.00008

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@inproceedings{6aed0932823f426caf33ad27c05346e1,

title = "360ViewPET: View Based Pose EsTimation for Ultra-Sparse 360-Degree Cameras",

abstract = "Immersive virtual tours based on 360-degree cameras, showing famous outdoor scenery, are becoming more and more desirable due to travel costs, pandemics and other constraints. To feel immersive, a user must receive the view accurately corresponding to her position and orientation in the virtual space when she moves inside, and this requires cameras' orientations to be known. Outdoor tour contexts have numerous, ultra-sparse cameras deployed across a wide area, making camera pose estimation challenging. As a result, pose estimation techniques like SLAM, which require mobile or dense cameras, are not applicable. In this paper we present a novel strategy called 360ViewPET, which automatically estimates the relative poses of two stationary, ultra-sparse (15 meters apart) 360-degree cameras using one equirectangular image taken by each camera. Our experiments show that it achieves accurate pose estimation, with a mean error as low as 0.9 degree. {\textcopyright} 2021 IEEE.",

keywords = "360 camera, pose estimation, virtual tourism",

author = "Qian Zhou and Bo Chen and Zhe Yang and Hongpeng Guo and Klara Nahrstedt",

year = "2021",

month = nov,

doi = "10.1109/ISM52913.2021.00008",

language = "English",

isbn = "9781665437356",

series = "Proceedings - IEEE International Symposium on Multimedia, ISM",

publisher = "IEEE",

pages = "1--8",

booktitle = "Proceedings - 23rd IEEE International Symposium on Multimedia, ISM 2021",

address = "United States",

note = "23rd IEEE International Symposium on Multimedia (ISM 2021), ISM2021 ; Conference date: 29-11-2021 Through 01-12-2021",

url = "https://semanticcomputing.wixsite.com/website-15",

}

Zhou, Q, Chen, B, Yang, Z, Guo, H & Nahrstedt, K 2021, 360ViewPET: View Based Pose EsTimation for Ultra-Sparse 360-Degree Cameras. in Proceedings - 23rd IEEE International Symposium on Multimedia, ISM 2021. Proceedings - IEEE International Symposium on Multimedia, ISM, IEEE, Los Alamitos, Calif., pp. 1-8, 23rd IEEE International Symposium on Multimedia (ISM 2021), 29/11/21. https://doi.org/10.1109/ISM52913.2021.00008

360ViewPET: View Based Pose EsTimation for Ultra-Sparse 360-Degree Cameras. / Zhou, Qian; Chen, Bo; Yang, Zhe et al.
Proceedings - 23rd IEEE International Symposium on Multimedia, ISM 2021. Los Alamitos, Calif.: IEEE, 2021. p. 1-8 (Proceedings - IEEE International Symposium on Multimedia, ISM).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - 360ViewPET

T2 - 23rd IEEE International Symposium on Multimedia (ISM 2021)

AU - Zhou, Qian

AU - Chen, Bo

AU - Yang, Zhe

AU - Guo, Hongpeng

AU - Nahrstedt, Klara

PY - 2021/11

Y1 - 2021/11

N2 - Immersive virtual tours based on 360-degree cameras, showing famous outdoor scenery, are becoming more and more desirable due to travel costs, pandemics and other constraints. To feel immersive, a user must receive the view accurately corresponding to her position and orientation in the virtual space when she moves inside, and this requires cameras' orientations to be known. Outdoor tour contexts have numerous, ultra-sparse cameras deployed across a wide area, making camera pose estimation challenging. As a result, pose estimation techniques like SLAM, which require mobile or dense cameras, are not applicable. In this paper we present a novel strategy called 360ViewPET, which automatically estimates the relative poses of two stationary, ultra-sparse (15 meters apart) 360-degree cameras using one equirectangular image taken by each camera. Our experiments show that it achieves accurate pose estimation, with a mean error as low as 0.9 degree. © 2021 IEEE.

AB - Immersive virtual tours based on 360-degree cameras, showing famous outdoor scenery, are becoming more and more desirable due to travel costs, pandemics and other constraints. To feel immersive, a user must receive the view accurately corresponding to her position and orientation in the virtual space when she moves inside, and this requires cameras' orientations to be known. Outdoor tour contexts have numerous, ultra-sparse cameras deployed across a wide area, making camera pose estimation challenging. As a result, pose estimation techniques like SLAM, which require mobile or dense cameras, are not applicable. In this paper we present a novel strategy called 360ViewPET, which automatically estimates the relative poses of two stationary, ultra-sparse (15 meters apart) 360-degree cameras using one equirectangular image taken by each camera. Our experiments show that it achieves accurate pose estimation, with a mean error as low as 0.9 degree. © 2021 IEEE.

KW - 360 camera

KW - pose estimation

KW - virtual tourism

UR - http://www.scopus.com/inward/record.url?scp=85125010686&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85125010686&origin=recordpage

U2 - 10.1109/ISM52913.2021.00008

DO - 10.1109/ISM52913.2021.00008

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 9781665437356

T3 - Proceedings - IEEE International Symposium on Multimedia, ISM

SP - 1

EP - 8

BT - Proceedings - 23rd IEEE International Symposium on Multimedia, ISM 2021

PB - IEEE

CY - Los Alamitos, Calif.

Y2 - 29 November 2021 through 1 December 2021

ER -

360ViewPET: View Based Pose EsTimation for Ultra-Sparse 360-Degree Cameras

Abstract

Publication series

Conference

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this