中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 自动化研究所
  3. 中国科学院自动化研究所
  4. 多模态人工智能系统全国重点实验室
  5. 机器人视觉团队
VidSfM: Robust and Accurate Structure-From-Motion for Monocular Videos

文献类型：期刊论文

作者Cui, Hainan3,4; Tu, Diantao3,4; Tang, Fulin3,4; Xu, Pengfei2; Liu, Hongmin1; Shen, Shuhan3,4
刊名IEEE TRANSACTIONS ON IMAGE PROCESSING
出版日期2022
卷号31页码:2449-2462
关键词Cameras Image reconstruction Videos Simultaneous localization and mapping Video sequences Robustness Scalability Structure from motion image reconstruction computational geometry computer vision
ISSN号1057-7149
DOI10.1109/TIP.2022.3156375
通讯作者Liu, Hongmin(hmliu_82@163.com) ; Shen, Shuhan(shshen@nlpr.ia.ac.cn)
英文摘要With the popularization of smartphones, larger collection of videos with high quality is available, which makes the scale of scene reconstruction increase dramatically. However, high-resolution video produces more match outliers, and high frame rate video brings more redundant images. To solve these problems, a tailor-made framework is proposed to realize an accurate and robust structure-from-motion based on monocular videos. The key ideas include two points: one is to use the spatial and temporal continuity of video sequences to improve the accuracy and robustness of reconstruction; the other is to use the redundancy of video sequences to improve the efficiency and scalability of system. Our technical contributions include an adaptive way to identify accurate loop matching pairs, a cluster-based camera registration algorithm, a local rotation averaging scheme to verify the pose estimate and a local images extension strategy to reboot the incremental reconstruction. In addition, our system can integrate data from different video sequences, allowing multiple videos to be simultaneously reconstructed. Extensive experiments on both indoor and outdoor monocular videos demonstrate that our method outperforms the state-of-the-art approaches in robustness, accuracy and scalability.
WOS关键词EFFICIENT ; VERSATILE
资助项目National Natural Science Foundation of China[62073320] ; National Natural Science Foundation of China[U1805264] ; National Natural Science Foundation of China[61873265] ; Didi Gaia Foundation
WOS研究方向Computer Science ; Engineering
语种英语
WOS记录号WOS:000769973200010
出版者IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
资助机构National Natural Science Foundation of China ; Didi Gaia Foundation
源URL[http://ir.ia.ac.cn/handle/173211/48129]  
专题自动化研究所_模式识别国家重点实验室_机器人视觉团队
通讯作者Liu, Hongmin; Shen, Shuhan
作者单位1.Univ Sci & Technol, Sch Automat & Elect Engn, Beijing 100183, Peoples R China
2.Didi Chuxing Technol Co, Beijing 100193, Peoples R China
3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100149, Peoples R China
4.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China
推荐引用方式
GB/T 7714		 Cui, Hainan,Tu, Diantao,Tang, Fulin,et al. VidSfM: Robust and Accurate Structure-From-Motion for Monocular Videos[J]. IEEE TRANSACTIONS ON IMAGE PROCESSING,2022,31:2449-2462.
APA Cui, Hainan,Tu, Diantao,Tang, Fulin,Xu, Pengfei,Liu, Hongmin,&Shen, Shuhan.(2022).VidSfM: Robust and Accurate Structure-From-Motion for Monocular Videos.IEEE TRANSACTIONS ON IMAGE PROCESSING,31,2449-2462.
MLA Cui, Hainan,et al."VidSfM: Robust and Accurate Structure-From-Motion for Monocular Videos".IEEE TRANSACTIONS ON IMAGE PROCESSING 31(2022):2449-2462.

入库方式： OAI收割

来源：自动化研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。