Presentation 2017-06-23
Global Metric Localization and Correction of Scale Drift using Street View
Kazuya Iwami, Toshihiko Yamasaki, Kiyoharu Aizawa,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper, we propose a novel method for video geo-localization using geo-tagged images, such as those retrieved from Google Street View. In the proposed method, correspondences of 3D positions between video frames and geo-tagged images are obtained, and then a reconstructed 3D map from input video is transformed to the world coordinates using the correspondences. However, applying a linear transformation to the 3D map leads to inaccuracy of geo-localization, because 3D reconstruction from a video has a problem that scale errors are accumulated in reconstruction---scale drift problem. To deal with this problem, we propose a transformation method that consists of three coarse-to-fine steps. Experiments show that the proposed method results accurate localization with errors of approximately 5~m on videos captured in more than 1~km trajectories, and the proposed method outperforms other state-of-the-art geo-localization methods that use geo-tagged images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) 3D Reconstruction / SLAM / Localization / Geo-Tag
Paper # PRMU2017-37,SP2017-13
Date of Issue 2017-06-15 (PRMU, SP)

Conference Information
Committee PRMU / SP
Conference Date 2017/6/22(2days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Shinichi Sato(NII) / Yoichi Yamashita(Ritsumeikan Univ.)
Vice Chair Hironobu Fujiyoshi(Chubu Univ.) / Yoshihisa Ijiri(Omron) / Hiroki Mori(Utsunomiya Univ.)
Secretary Hironobu Fujiyoshi(AIST) / Yoshihisa Ijiri(NAIST) / Hiroki Mori(Shizuoka Univ.)
Assistant Masato Ishii(NEC) / Yusuke Sugano(Osaka Univ.) / Kei Hashimoto(Nagoya Inst. of Tech.) / Satoshi Kobashikawa(NTT)

Paper Information
Registration To Technical Committee on Pattern Recognition and Media Understanding / Technical Committee on Speech
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Global Metric Localization and Correction of Scale Drift using Street View
Sub Title (in English)
Keyword(1) 3D Reconstruction
Keyword(2) SLAM
Keyword(3) Localization
Keyword(4) Geo-Tag
1st Author's Name Kazuya Iwami
1st Author's Affiliation The University of Tokyo(Univ. of Tokyo)
2nd Author's Name Toshihiko Yamasaki
2nd Author's Affiliation The University of Tokyo(Univ. of Tokyo)
3rd Author's Name Kiyoharu Aizawa
3rd Author's Affiliation The University of Tokyo(Univ. of Tokyo)
Date 2017-06-23
Paper # PRMU2017-37,SP2017-13
Volume (vol) vol.117
Number (no) PRMU-105,SP-106
Page pp.pp.69-74(PRMU), pp.69-74(SP),
#Pages 6
Date of Issue 2017-06-15 (PRMU, SP)