idanpa / SURFMNet-Multispectral

Source code for the paper "Unsupervised Scale-Invariant Multispectral Shape Matching"

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

SURFMNet-Multispectral

Source code for the paper "Unsupervised Scale-Invariant Multispectral Shape Matching" by Idan Pazi, Dvir Ginzburg and Dan Raviv.
Based on the source code of SURFMNet.

Shape Matching

To train a model

    python train_DFMnet.py

To obtain matches after training for a given set of shapes:

    python test_DFMnet.py

About

Source code for the paper "Unsupervised Scale-Invariant Multispectral Shape Matching"


Languages

Language:Python 92.3%Language:Perl 6.0%Language:Shell 1.7%