CASIA OpenIR  > 模式识别国家重点实验室  > 机器人视觉
Robust global translation averaging with feature tracks
Cui Hainan(崔海楠); Shen Shuhan(申抒含); Hu Zhanyi(胡占义)
2016
Conference NameInternational Conference on Pattern Recognition
Conference Date2016-11
Conference PlaceMexico
AbstractHow to average translations is the single most difficult task in global structure-from-motion (SfM) to fully tap its potentials in terms of reconstruction efficiency and accuracy since usually only noisy translation directions can be factored out from essential matrices due to the inevitable matching outliers. To tackle this problem, this work proposes a two-step strategy. Firstly, a “2-point method” is introduced to refine the epipolar geometry by which a more accurate track set is generated. Then, translation lengths are computed by solving a convex L1 optimization according to the adjacent triangles induced by the selected tracks and translations. Extensive experiments show that our method performs similarly or better than the state-of-art SfM approaches in terms of the reconstruction accuracy, completeness and efficiency. 
Document Type会议论文
Identifierhttp://ir.ia.ac.cn/handle/173211/19767
Collection模式识别国家重点实验室_机器人视觉
Recommended Citation
GB/T 7714
Cui Hainan,Shen Shuhan,Hu Zhanyi. Robust global translation averaging with feature tracks[C],2016.
Files in This Item: Download All
File Name/Size DocType Version Access License
HainanCui_ICPR2016.p(1224KB)会议论文 开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Cui Hainan(崔海楠)]'s Articles
[Shen Shuhan(申抒含)]'s Articles
[Hu Zhanyi(胡占义)]'s Articles
Baidu academic
Similar articles in Baidu academic
[Cui Hainan(崔海楠)]'s Articles
[Shen Shuhan(申抒含)]'s Articles
[Hu Zhanyi(胡占义)]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Cui Hainan(崔海楠)]'s Articles
[Shen Shuhan(申抒含)]'s Articles
[Hu Zhanyi(胡占义)]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: HainanCui_ICPR2016.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.