Knowledge Network Node
Modern Computer. 2018, (08)

An Improved Local Geometric Feature Matching Method for Point CloudsChinese Full Text

SHI Kui-yang;College of Computer Science,Sichuan University;State Key Laboratory of Fundamental Science on Synthetic Vision,Sichuan University;

Abstract: The establishment of correspondence between 3D objects through the description of feature points is of great significance in various graphic and visual applications such as 3D reconstruction,location and shape matching.For general descriptor,it is difficult to deal with the noise,low resolution and incomplete information of the 3D point cloud data in the real world,3D Match propose a method based on Convolutional Neural Network,learning a new local volume descriptor,and the descriptor is applied... More
Keywords：

Convolutional Neural Network; 3D Match; Local Feature Descriptor; Point Cloud;

  • Series：

    (I) Electronic Technology ＆ Information Science

  • Subject：

    Computer Software and Application of Computer

  • Classification Code：

    TP391.41

Download the mobile appuse the app to scan this coderead the article.

Tips: Please download CAJViewer to view CAJ format full text.

Download: 156 Page: 61-65 Pagecount: 5 Size: 2068K

Related Literature
  • Similar Article
  • Reader Recommendation
  • Associated Author