Skip to content

Latest commit

 

History

History
30 lines (25 loc) · 1.51 KB

README.md

File metadata and controls

30 lines (25 loc) · 1.51 KB
Raw

Zero-Assignment Constraint for Graph Matching with Outliers

Implementation for the CVPR2020 paper.

Supplementary materials

Abstract

This work focuses on the problem of graph matching in the presence of outliers, based on the zero-assignment constraint. This repository consists of the code of our work ZAC/ZACR (ZAC with regularization) and the compared graph matching and point registration methods.

Details

  • Download the data from Google Drive, then unzip it to replace the "data/" folder existed.
  • The compared graph matching methods and point registration methods can be found in directories "src/GM_methods" and "src/PR_methods", respectively.
  • Run the demos to see some examples on the used datasets.

Citation

If you find our work useful please consider citing:

@inproceedings{ZAC,
title = "Zero-Assignment Constraint for Graph Matching with Outliers",
author = "Fudong Wang and Nan Xue and Jin-Gang Yu and Gui-Song Xia",
booktitle = "IEEE Conference on Computer Vision and Pattern Recognition (CVPR)",
year = {2020},
}