按:   检索词:   从
   最新录用    过刊检索
三维重建算法综述
投稿时间:2019-11-20  修订日期:2020-04-22  点此下载全文
引用本文:
摘要点击次数: 77
全文下载次数: 0
作者单位E-mail
张彦雯 南京信息工程大学 526531735@qq.com 
胡凯 南京信息工程大学 nuistpanda@163.com 
王鹏盛 南京信息工程大学  
基金项目:国家自然科学基金(61773219, 61701244),国家重点研发计划重点专项课题(2018YFC1405703)
中文摘要:三维重建技术能够应用在医学中的三维CT图像、机器人中的路径规划及考古工作中的遗迹的展示等诸多领域中。三维重建效果的完整性、精确性与稀疏点云重建(Structure From Motion,SFM)、稠密点云重建(Multi view System,MVS)有着直接关系,本文对这两个步骤展开了具体的介绍。稀疏点云重建中主要介绍了特征点检测与匹配以及SFM重构方法,本文对近几年关于特征点检测与匹配的研究进行了总结,将SFM重构方法分为全局式、增量式,并进行了详细介绍。稠密点云重建中主要介绍了深度图估计,就传统的几何计算法、基于深度学习的方法与传统方法相结合的方法、基于深度学习的方法进行了总结,同时介绍了三维重建的评价指标。最后对基于图像的三维重建进行了总结,并列举出了在未来发展中面对的问题及发展趋势。
中文关键词:三维重建  图像  稀疏点云重建  稠密点云重建  评价指标
 
A review of 3d reconstruction algorithms
Abstract:Three-dimensional reconstruction technology can be applied in many fields such as three-dimensional CT images in medicine, path planning in robots and display of relics in archaeological work.The completeness and accuracy of 3d reconstruction effect are directly related to Structure From Motion (SFM) and Multi-View System (MVS). This paper introduces these two steps in detail.In the reconstruction of sparse point cloud, feature point detection and matching and SFM reconstruction methods are mainly introduced. This paper summarizes the research on feature point detection and matching in recent years, and divides the reconstruction methods of SFM into global and incremental ones, and introduces them in detail.In the dense point cloud reconstruction, the depth map estimation is mainly introduced, and the traditional geometric calculation method, the method based on deep learning combined with the traditional method, and the method based on deep learning are summarized. At the same time, the evaluation index of 3d reconstruction is introduced.Finally, the 3d reconstruction based on image is summarized and the problems and trends in the future development are listed.
keywords:3D reconstruction  Images  SFM  MVS  The evaluation index
  查看/发表评论  下载PDF阅读器

您是本站第 1744636 位访问者
版权所有:南京信息工程大学期刊社《南京信息工程大学学报》编辑部     
地址:江苏南京,宁六路219号,南京信息工程大学