Skip to content

C++ implementation of the GMS Feature Correspondence Algorithm

Notifications You must be signed in to change notification settings

germain-hug/GMS-Feature-Correspondence

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

43 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature Correspondence

C++ implementation of the Grid-based Motion Statistics for Feature Correspondance, by Bian et al.

The Grid-Based Motion Statistics is a method for improving the quality of feature matches, through motion-smoothness encapsulation. Based on ORB feature matches, the smoothness constraint is expressed as a statistical likelihood of the per-cell matching frequency. Results show both real-time and very robust feature correspondence.

ORB feature points (2000) After GMS ORB Feature points (2000) After GMS

Running

This implementation requires OpenCV 2.4.13.2

mkdir build && make    
./GMS <path_to_image_1> <path_to_image_2>