Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Code and models for "Learning to Compare Image Patches via Convolutional Neural Networks"
| Date | Stars |
|---|---|
| 2026-07-31 | 472 |
| 2026-08-01 | 472 |
| 2026-08-02 | 472 |
| 2026-08-06 | 472 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
Code for CVPR15 paper "Learning to Compare Image Patches via Convolutional Neural Networks"
-----
This package allows researches to apply the described networks to match image patches and extract corresponding patches.
We tried to make the code as easy to use as possible. The original models were trained with Torch ( http://torch.ch ) and we release them in Torch7 and binary formats with C++ bindings which do not require Torch installation. Thus we provide example code how to use the models in Torch, MATLAB and with OpenCV http://opencv.org
CREDITS, LICENSE, CITATION
-----
Copyright © 2015 Ecole des Ponts, Universite Paris-Est
All Rights Reserved. A license to use and copy this software and its documentation solely for your internal
research and evaluation purposes, without fee and without a signed licensing agreement, is hereby granted
upon your download of the software, through which you agree to the following:
1) the above copyright notice, this paragraph and the following three paragraphs will prominently appear
in all internal copies and modifications;
2) no rights to sublicense or further distribute this software are granted;
3) no rights to modify this software are granted; and
4) no rights to assign this license are granted.
Please Contact Prof. Nikos Komodakis,
6 Avenue Blaise Pascal - Cite Descartes, Champs-sur-Marne, 77455 Marne-la-Vallee cedex 2, France for commercial licensing opportunities, or for further distribution, modification or license rights.
Created by Sergey Zagoruyko and Nikos Komodakis. http://imagine.enpc.fr/~komodakn/
Please cite the paper below if you use this code in your research.
Sergey Zagoruyko, Nikos Komodakis,
"Learning to Compare Image Patches via Convolutional Neural Networks". http://www.cv-foundation.org/openaccess/content_cvpr_2015/papers/Zagoruyko_Learning_to_Compare_2015_CVPR_paper.pdf, bib:
```
@InProceedings{Zagoruyko_2015_CVPR,
author = {Zagoruyko, Sergey and Komodakis, Nikos},
title = {Learning to Compare Image Patches via Convolutional Neural Networks},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2015}
}
```
**Update 4** (April 2017) dead links for models and datasets fixed
**Update 3** (July 2016) training code released
**Update 2** (February 2016) caffe models released
**Update 1** (January 2016) cudnn models removed because `cudnn.convert` was out
## Dataset
The original dataset website is down, you can still download the files here:
<http://icvl.ee.ic.ac.uk/vbalnt/notredame.zip><br>
<http://icvl.ee.ic.ac.uk/vbalnt/yosemite.zip><br>
<http://icvl.ee.ic.ac.uk/vbalnt/liberty.zip><br>
### Models
We provide the models in Torch7 and binary format. The table from the paper is here for convenience.
**All models expect input patches to be in [0;1] range before mean subtraction.**
**The models are not supposed to give outputs in [0;1] range, the outputs are not normalized**
| Train set | Test set | 2ch | 2ch2stream | 2chdeep | siam | siam2stream |
| --- | --- | :---: | :---: | :---: | :---: | :---: |
| yosemite | notredame | 2.74 | **2.11** | 2.43 | 5.62 | 5.23 |
| yosemite | liberty | 8.59 | **7.2** | 7.4 | 13.48 | 11.34 |
| notredame | yosemite | 6.04 | **4.09** | 4.38 | 13.23 | 10.44 |
| notredame | liberty | 6.04 | 4.85 | **4.56** | 8.77 | 6.45 |
| liberty | yosemite | 7 | **5** | 6.18 | 14.76 | 9.39 |
| liberty | notredame | 2.76 | **1.9** | 2.77 | 4.04 | 2.82 |
Models in nn format can be loaded and used without CUDA support in Torch. To enable CUDA support ```model:cuda()``` call required.
An archive with all models (binary and in torch format) is available at: <https://s3.amazonaws.com/modelzoo-networks/cvpr2015matching_networks.tar.gz>
### Torch
To install torch follow http://torch.ch/
Check torch folder for examples.
Match patches on CPU:
```lua
require 'nn'
N = 76 -- the number of patches to match
patches = torch.rand(N,2,64,64):float()
-- load the network
net = torch.Excerpt of 9,112 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:89ff2ed7c8d3bb1d, llm:Repository description: 'Code and models for "Learning to Compare Image Patches via Convolutional Neural Networks"' (C++).
matched fp:89ff2ed7c8d3bb1d, llm:Repository description: 'Code and models for "Learning to Compare Image Patches via Convolutional Neural Networks"' (C++).
matched fp:89ff2ed7c8d3bb1d, llm:Repository description: 'Code and models for "Learning to Compare Image Patches via Convolutional Neural Networks"' (C++).