# DRT

**Repository Path**: greitzmann/DRT

## Basic Information

- **Project Name**: DRT
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2019-11-14
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README

### Usage

* Supported OS: the source code was tested on 64-bit Arch and Ubuntu 14.04 Linux OS, and it should also be executable in other linux distributions.

* Dependencies: 
 * A modified version of [caffe](http://caffe.berkeleyvision.org/) framework and all its dependencies. 
 * Cuda enabled GPUs

* Installation: 
 1. Install caffe: we use a modified version of the original caffe framework. Compile the source code in the ./caffe directory and the matlab interface following the [installation instruction of caffe](http://caffe.berkeleyvision.org/installation.html).
 2. Download the 16-layer VGG network from https://gist.github.com/ksimonyan/211839e770f7b538e2d8, and put the caffemodel file under the ./model directory.
 3. Download imagenet-vgg-m-2048 from http://www.vlfeat.org/matconvnet/pretrained/, and put the file into ./networks
 4. Compile matconvnet in the sub-folders.
 5. Run the demo code demo_DRT.m. You can customize your own test sequences following this example.

The tracking results may be a little different on different machines. The suggested MATLAB and CUDA versions are MATLAB R2014B and CUDA 8.0. 

If you find our paper useful, please consider citing it.
@inproceedings{sun2018correlation,
  title={Correlation Tracking via Joint Discrimination and Reliability Learning},
  author={Sun, Chong and Wang, Dong and Lu, Huchuan and Yang, Ming-Hsuan},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  pages={489--497},
  year={2018}
}