###
Journal of Software:2016.27(11):2929-2945

基于自然场景在线学习的跟踪注册技术
桂振文,刘越,陈靖,王涌天
(中国电子科技集团公司 第七研究所, 广东 广州 510310;北京理工大学 光电学院, 北京 100081;北京市混合现实与新型显示工程技术研究中心(北京理工大学), 北京 100081)
Online Learning of Tracking and Registration Based on Natural Scenes
GUI Zhen-Wen,LIU Yue,CHEN Jing,WANG Yong-Tian
(No. 7 Research Institute, China Electronics Technology Group Corporation, Guangzhou 510310, China;School of Optoelectronics, Beijing Institute of Technology, Beijing 100084, China;Beijing Engineering Research Center for Mixed Reality and Advanced Display(Beijing Institute of Technology, Beijing 100081, China)
Abstract
Chart / table
Reference
Similar Articles
Article :Browse 1276   Download 1125
Received:March 04, 2014    Revised:June 28, 2014
> 中文摘要: 三维注册是移动增强现实的关键技术之一,提出了一种在线学习的跟踪注册方法,能够精确地对自然场景进行跟踪注册.该方法首先改进SURF(speeded up robust features)描述符匹配方法,提高初始注册矩阵的正确性;然后,通过对场景进行有效的在线学习,提高注册精度;最后,利用前一帧的注册矩阵快速恢复已丢失的关键点,以提高注册的速度.实验结果表明,该方法能够较为流畅地对视频帧进行跟踪,并能保持较好的注册精度.
Abstract:Registration is a fundamental technology for augmented reality. In this paper, a registration approach is proposed to accurately track the natural scenes. The matching method of SURF (speeded up robust features) descriptor is first improved to keep the initial registration matrix validity. Then, effective online learning of the scenes is used to improve the registration accuracy. Lastly, the registration matrix of the previous frame is utilized to rapidly restore the lost key points and accelerate the speed of registration. Experimental results show that the proposed method can keep smooth tracking for video frames and maintain high accuracy of registration.
文章编号:     中图分类号:    文献标志码:
基金项目:国家自然科学基金(61072096,60903070);国家高技术研究发展计划(863)(2013AA013802);国家“十二五”重点科技攻关项目(2012ZX03002004);广东省协同创新与平台环境建设专项(2014B090901024) 国家自然科学基金(61072096,60903070);国家高技术研究发展计划(863)(2013AA013802);国家“十二五”重点科技攻关项目(2012ZX03002004);广东省协同创新与平台环境建设专项(2014B090901024)
Foundation items:National Natural Science Foundation of China (61072096, 60903070); National High-Tech R&D Program of China (863) (2013AA013802); Key Science-Technology Project of the National ‘Twelfth Five-Year-Plan’ of China (2012ZX03002004); Collaborative Innovation and Platform Environment Construction Major Project of Guangdong Province (2014B090901024)
Reference text:

桂振文,刘越,陈靖,王涌天.基于自然场景在线学习的跟踪注册技术.软件学报,2016,27(11):2929-2945

GUI Zhen-Wen,LIU Yue,CHEN Jing,WANG Yong-Tian.Online Learning of Tracking and Registration Based on Natural Scenes.Journal of Software,2016,27(11):2929-2945