Skip to content

Matlab code for the paper "Using Spatial Order to Boost the Elimination of Incorrect Feature Matches", CVPR 2016, Lior Talker, Yael Moses and Ilan Shimshoni

Notifications You must be signed in to change notification settings

liortalker/SpatialOrder

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SpatialOrder

This is code provides a minimal working (MWE) example of the paper "Using Spatial Order to Boost the Elimination of Incorrect Feature Matches" by Lior Talker, Yael Moses and Ilan Shimshoni, presented at CVPR 2016. For questions or information about the paper or code please contact Lior Talker (liortalker@gmail.com or http://liortalker.wix.com/liortalker) Only for academic or other non-commercial purposes (under GPL terms).

Running the algorithm:

For the synthetic experiments run SyntheticExp1.m or SyntheticExp2.m (Provide only the number of matches and (for Exp1 the number of inliers)

For a real experiments run either runKendallInlierRate.m (no parameters - loads im1.jpg and im2.jpg), or runKendallInlierRateOnTwoImages.m for which the input is two images and parameters, or estimateKendallInlierRateSeparateWindows.m (or estimateKendallInlierRateJointlyWindows.m) for which the input is two sequences and a resolution parameter usually = 0.1)

Dependencies:

The algorithm itself has no dependencies, but to extract features (e.g., SIFT) the example is dependent on VLFeat. You can download it from http://www.vlfeat.org, place it in a folder inside the project, invoke vl_setup, and run the example itself.

About

Matlab code for the paper "Using Spatial Order to Boost the Elimination of Incorrect Feature Matches", CVPR 2016, Lior Talker, Yael Moses and Ilan Shimshoni

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published