https://github.com/essishen93/Product-Quantilization
The implementation of the paper " Product quantization for nearest neighbor search"
https://github.com/essishen93/Product-Quantilization
Last synced: 27 days ago
JSON representation
The implementation of the paper " Product quantization for nearest neighbor search"
- Host: GitHub
- URL: https://github.com/essishen93/Product-Quantilization
- Owner: essishen93
- Created: 2016-11-11T04:21:41.000Z (over 8 years ago)
- Default Branch: master
- Last Pushed: 2016-11-11T04:36:34.000Z (over 8 years ago)
- Last Synced: 2024-10-27T20:18:49.396Z (6 months ago)
- Language: C++
- Size: 24.4 KB
- Stars: 11
- Watchers: 0
- Forks: 5
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- my-awesome-awesomeness - Product-Quantization
README
# Product-Quantilization
The implementation of the paper " Product quantization for nearest neighbor search"
These codes are the modified version of codes from the following author:
https://github.com/lostmarble/product-quantization)
/**
* @file config_file.h
* @Synopsis read configure file
* @author [email protected]
* @version 0.1
* @date 2012-06-02
*/
This project contains all the tools to find stable points in a
picture.Additionally, it can also be used as an picture search
engine.
The project mainly impletes the algorithm introduced by:
"Product quantization for nearest neighbor search"
Hervé Jégou, Matthijs Douze and Cordelia Schmid, 2011 TPAMI.
and also use a library(yael) from their project, all rights own to them.
1. Feature Extraction
Run "extracter picture_dir feature_dir"
picture_dir: the directory storing the original images
feature_dir: the directory storing the output feature files
2.Train and build the search dataset
Run "pqtrain base_pic_feature_dir"
output files: model.dat, database.dat
3. Query a image
Run "pqsearch query_pic_feature_dir/query_pic_feature_list_file"
The program will aotumatically load model.dat and database.dat to query