Presentation 2022-02-22
Object Detection in 360-degree images Based on Multiple Equirectangular Projections
Cao Miao, Satoshi Ikehata, Kiyoharu Aizawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) 360° cameras have gained popularity over the last few years. Meanwhile, the development of Deep Neural Networks (DNNs) has dramatically improved the accuracy of object detection. However, the accuracy of object detection in 360-degree images has not been satisfactory due to the distortion of the equirectangular projection (ERP) images. In this paper, we propose a multi-view ERP representation as input for the training and inference of object detectors. In addition, we propose a data augmentation technique for object detection in 360° images. The proposed augmentation technique is specific to the 360° object detection task which randomly rotates the spherical image to improve the robustness against ERP image distortion. The proposed methods can be readily integrated into existing object detectors and significantly improve the performance. The experimental results show the significant effect of our method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) 360-degree images / object detection
Paper # ITS2021-51,IE2021-60
Date of Issue 2022-02-14 (ITS, IE)

Conference Information
Committee IE / ITS / ITE-AIT / ITE-ME / ITE-MMS
Conference Date 2022/2/21(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Online
Topics (in Japanese) (See Japanese page)
Topics (in English) Image Processing, etc.
Chair Kazuya Kodama(NII) / Masahiro Fujii(Utsunomiya Univ.) / Hisaki Nate(Tokyo Polytechnic Univ.) / Hiroyuki Arai(Nippon Inst. of Tech.) / Kenji Machida(NHK)
Vice Chair Hiroyuki Bandoh(NTT) / Toshihiko Yamazaki(Univ. of Tokyo) / Kohei Ohno(Meiji Univ.) / Naohisa Hashimoto(AIST) / / Shogo Muramatsu(Niigata Univ.)
Secretary Hiroyuki Bandoh(KDDI Research) / Toshihiko Yamazaki(Nagoya Inst. of Tech.) / Kohei Ohno(Akita Prefectural Univ.) / Naohisa Hashimoto(NIT, Tsuruoka College) / / Shogo Muramatsu(NHK) / (Hokkaido Univ.)
Assistant Shunsuke Iwamura(NHK) / Shinobu Kudo(NTT) / Msataka Imao(Mitsubishi Electric) / Kenshi Saho(Toyama Prefectural Univ.) / Keiji Jimi(Gunma Univ.)

Paper Information
Registration To Technical Committee on Image Engineering / Technical Committee on Intelligent Transport Systems Technology / Technical Group on Artistic Image Technology / Technical Group on Media Engineering / Technical Group on Multi-media Storage
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Object Detection in 360-degree images Based on Multiple Equirectangular Projections
Sub Title (in English)
Keyword(1) 360-degree images
Keyword(2) object detection
1st Author's Name Cao Miao
1st Author's Affiliation The University of Tokyo(UTokyo)
2nd Author's Name Satoshi Ikehata
2nd Author's Affiliation National Institute of Informatics(NII)
3rd Author's Name Kiyoharu Aizawa
3rd Author's Affiliation The University of Tokyo(UTokyo)
Date 2022-02-22
Paper # ITS2021-51,IE2021-60
Volume (vol) vol.121
Number (no) ITS-373,IE-374
Page pp.pp.157-162(ITS), pp.157-162(IE),
#Pages 6
Date of Issue 2022-02-14 (ITS, IE)