https://github.com/JiawangBian/GMS-Feature-Matcher
Revision 2c285e32059d35fe19eeeb7ba739b4e8160b82fe authored by JiaWang Bian on 06 April 2020, 02:07:39 UTC, committed by GitHub on 06 April 2020, 02:07:39 UTC
1 parent 0f2f654
Tip revision: 2c285e32059d35fe19eeeb7ba739b4e8160b82fe authored by JiaWang Bian on 06 April 2020, 02:07:39 UTC
Update README.md
Update README.md
Tip revision: 2c285e3
README.md
# GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature Correspondence
![alt tag](http://mmcheng.net/wp-content/uploads/2017/03/dog_ours.jpg)
## Publication:
[JiaWang Bian](http://jwbian.net), Wen-Yan Lin, Yasuyuki Matsushita, Sai-Kit Yeung, Tan Dat Nguyen, Ming-Ming Cheng, **GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature Correspondence**, **CVPR 2017**, [[Project Page](http://jwbian.net/gms)] [[pdf](http://jwbian.net/Papers/GMS_CVPR17.pdf)] [[Bib](http://jwbian.net/Papers/bian2017gms.txt)] [[Code](https://github.com/JiawangBian/GMS-Feature-Matcher)] [[Youtube](https://youtu.be/3SlBqspLbxI)]
[JiaWang Bian](http://jwbian.net), Wen-Yan Lin, Yun Liu, Le Zhang, Sai-Kit Yeung, Ming-Ming Cheng, Ian Reid, **GMS: Grid-based Motion Statistics for Fast, Ultra-robust Feature Correspondence**, **IJCV 2019**, [[pdf](https://link.springer.com/content/pdf/10.1007%2Fs11263-019-01280-3.pdf)]
## Other Resouces
The method has been integrated into OpenCV library (see [xfeatures2d.matchGMS](https://docs.opencv.org/master/db/dd9/group__xfeatures2d__match.html)).
More experiments are shown in [FM-Bench](https://jwbian.net/fm-bench).
The paper was selected and reviewed by [Computer Vision News](http://www.rsipvision.com/ComputerVisionNews-2017August/#48).
## If you find this work useful in your research, please consider citing our paper:
@article{Bian2019gms,
title={{GMS}: Grid-based Motion Statistics for Fast, Ultra-Robust Feature Correspondence},
author={Bian, JiaWang and Lin, Wen-Yan and Liu, Yun and Zhang, Le and Yeung, Sai-Kit and Cheng, Ming-Ming and Reid, Ian},
journal={International Journal of Computer Vision (IJCV)},
year={2019}
}
## Usage
Requirement:
1.OpenCV 3.0 or later (for IO and ORB features, necessary)
2.cudafeatures2d module(for gpu nearest neighbor, optional)
C++ Example:
Image pair demo in demo.cpp.
Matlab Example
You should compile the code with opencv library firstly(see the 'Compile.m').
Python Example:
Use Python3 to run gms_matcher script.
Tune Parameters:
In demo.cpp
1. #define USE_GPU" (need cudafeatures2d module)
using cpu mode by commenting it.
2. We suggest using SIFT features for accuracy, and using ORB features for speed.
In gms_matcher.h
2. #define THRESH_FACTOR 6
Set it higher for more input matches, and lower for the fewer input matches.
Often 6 for ORB raw matches, and 4 or 3 for SIFT matches (after ratio test).
3. int GetInlierMask(vector<bool> &vbInliers, bool WithScale = false, bool WithRotation = false)
Set WithScale to be true for wide-baseline matching and false for video matching.
## Related projects
* [FM-Bench](https://github.com/JiawangBian/FM-Bench) (BMVC 2019, More evaluation details for GMS.)
Computing file changes ...