Skip to content

Latest commit

 

History

History
72 lines (48 loc) · 2.68 KB

README.md

File metadata and controls

72 lines (48 loc) · 2.68 KB
Raw

Triplet Binary Embeddings

The aim of this project is to form Human Pose Embeddings based on a strategy described in a paper "Fast Training of Triplet-based Deep Binary Embedding Networks by Bohan Zhuang, Guosheng Lin, Chunhua Shen, Ian Reid." http://arxiv.org/abs/1603.02844

Paper: "Fast Training of Triplet-based Deep Binary Embedding Networks"

##People Bohan Zhuang, Guosheng Lin, Chunhua Shen and Ian Reid. Code author: Bohan Zhuang This code is provided for non-profit research purpose only; and is released under the GNU license. For commercial applications, please contact Chunhua Shen http://www.cs.adelaide.edu.au/~chhshen/.

This is the implementation of the following paper. If you use this code in your research, please cite our paper

@InProceedings{Zhuang_2016_CVPR,
author = {Zhuang, Bohan and Lin, Guosheng and Shen, Chunhua and Reid, Ian},
title = {Fast Training of Triplet-Based Deep Binary Embedding Networks},
booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2016}
}

Overview

./step1/ includes the code for the binary codes inference step. ./lib/ includes the necessary codes for the network training. We inplement it using Theano. ./preprocessing is the data preprocessing toolbox.

Data Preprocessing

The data preprocessing code is in ./preprocessing/ and it will generate suitable data for training and testing. Please modify ./preprocessing/paths.yaml.

run make_caffe_txt.py-->make_hkl.py-->make_labels.py

The processed data is in folder ./preprocessed_data/.

Training

The code is based on Ubuntu 14.04. The main function is the ./step1/train.m file. Please modify the configurations in the config.yaml. Trained models will be stored in ./models/.

Testing

You should first extract the gallery codes as well as the query codes by running ./code_extraction.py. Then run ./test.m to do testing.

Copyright

Copyright (c) Bohan Zhuang. 2016.

** This code is for non-commercial purposes only. For commerical purposes, please contact Chunhua Shen chhshen@gmail.com **

This program is free software: you can redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation, either version 3 of the License, or (at your option) any later version.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for more details.

You should have received a copy of the GNU General Public License along with this program. If not, see http://www.gnu.org/licenses/.