• 设为首页
  • 点击收藏
  • 手机版
    手机扫一扫访问
    迪恩网络手机版
  • 关注官方公众号
    微信扫一扫关注
    迪恩网络公众号

pcirujeda/CovGT-3DRegistration-matlab: A 3D Scene Registration Method via Covari ...

原作者: [db:作者] 来自: 网络 收藏 邀请

开源软件名称(OpenSource Name):

pcirujeda/CovGT-3DRegistration-matlab

开源软件地址(OpenSource Url):

https://github.com/pcirujeda/CovGT-3DRegistration-matlab

开源编程语言(OpenSource Language):

MATLAB 100.0%

开源软件介绍(OpenSource Introduction):

A 3D Scene Registration Method via Covariance Descriptors and an Evolutionary Stable Strategy Game Theory Solver

About

This is a Matlab implementation of the method "A 3D Scene Registration Method via Covariance Descriptors and an Evolutionary Stable Strategy Game Theory Solver".

It provides a 3D point cloud descriptor for the local definition of points, fusing shape and color information, based on the notion of covariance of features. The intrinsic properties of this descriptor are many: it is invariant to spatial rigid transformations, and robust to noise and resolution changes; it can also be used for characteristic point detection; and lies on top of a manifold topology which allows the use of analytical metric properties.

The method is complemented with a Game Theory based approach for solving the matching correspondences under global geometric constraints. This offers a comprehensive understanding of the scene and avoids possible mismatches due to repeated areas or symmetries, which would be impossible to identify by the detector solely at a local level.

The full registration approach is able to accurately match different views of a scene under spatial transformations, high noise levels and with small overlap between views.

The method is described in detail in the following publications IJCV 2015, 3DV 2014. Please cite them if you use it in you research and/or applications:

@article{ IJCV2015,
  title={ A 3{D} scene registration method via covariance descriptors and an evolutionary stable strategy game theory solver },
  author={ Cirujeda, Pol and Cid, Yashin Dicente and Mateo, Xavier and Binefa, Xavier },
  journal={ International Journal of Computer Vision (IJCV) },
  volume={ 115 },
  number={ 3 },
  pages={ 306--329 },
  year={ 2015 },
  publisher={ Springer }
}
@inproceedings{ 3DV2014,
  title={ {MCOV}: a covariance descriptor for fusion of texture and shape features in 3{D} point clouds },
  author={ Cirujeda, Pol and Mateo, Xavier and Cid, Yashin Dicente and Binefa, Xavier },
  booktitle={ International Conference on 3D Vision (3DV), 2014 },
  volume={ 1 },
  pages={ 551--558 },
  year={ 2014 },
  organization={ IEEE }
}

Usage

The method is targeted at registration of 3D point clouds containing 3D coordinates, vertex normals and colors in RGB space, with provided support for reading and storing results to Wavefront OBJ files.

Please check the included runme.m script for a sample of a method call and parameterisation. All the methods included in this release are pretty much commented and self-explanatory.

A sample scene is included in the ./Samples folder. Otherwise you can check out the 3DVis dataset repository in order to get a set of 12 scenes as used in the experimental evaluation sections of the aforementioned publications.

If verbose mode is enabled you will be able to see the different steps of the registration procedure, including scene analysis for keypoint extraction, Covariance Descriptor likelihood matching, Game Theory correspondences calculation and final registration estimation.

License & Attribution

This software release is primarily MIT licensed. Some files contain third-party code under other licenses.




鲜花

握手

雷人

路过

鸡蛋
该文章已有0人参与评论

请发表评论

全部评论

专题导读
热门推荐
阅读排行榜

扫描微信二维码

查看手机版网站

随时了解更新最新资讯

139-2527-9053

在线客服(服务时间 9:00~18:00)

在线QQ客服
地址:深圳市南山区西丽大学城创智工业园
电邮:jeky_zhao#qq.com
移动电话:139-2527-9053

Powered by 互联科技 X3.4© 2001-2213 极客世界.|Sitemap