Skip to content

Latest commit

 

History

History
55 lines (35 loc) · 1.51 KB

README.md

File metadata and controls

55 lines (35 loc) · 1.51 KB
Raw

Source code for paper "Discrete Latent Factor Model for Cross-Modal Hashing"

Introduction

0. About the paper

This repo is the source code for the paper "Discrete Latent Factor Model for Cross-Modal Hashing" on TIP-2019. The authors are: Qing-Yuan Jiang and Wu-Jun Li. If you have any questions about the source code, please contact: qyjiang24#gmail.com.

We provide two versions for DLFH, i.e., matlab version and python version. Please see matlab_version and python_version folder , respectively.

1. Running Environment

For matlab version.

Matlab 2016

For python version.

python 3

2. Datasets

We use three datasets to perform our experiments, i.e., MIRFLICKR-25K, IAPR-TC12 and NUS-WIDE datasets.

2.1. Dataset download:

You can download all dataset from pan.baidu.com. The links are listed as follows:

3. Run demo

3.1. Matlab version:

Run matlab_version/DLFH_demo.m.

DLFH_demo

3.2. Python version:

Run python_version/dlfh_demo.py

python dlfh_demo.py --bit 8

Please note that we only present a simple example for python version. Hence we only implement DLFH for python version. All exmperiments in the paper are based on matlab version.