Robust and real-time deep tracking via multi-scale domain adaptation
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
model
utility
videos/Bolt
README.md
demo.m
run_tracker.m
tracker_ensemble.m

README.md

README

Network image

Overview

  • This is the test code of the Multi-Scale Domain Adaptation Tracker (MSDAT).
  • This code is relative to an arXiv tech report, which is accepted on ICME 2017.
  • Project web is here, where you can download the results.
  • Version - 0.1

Install

  • Install the Caffe deep learning toolbox with the MatCaffe interface
  • In demo.m, replace the "PATH_TO_CAFFE" in "addpath('PATH_TO_CAFFE/caffe-master/matlab/')" with the appropreate path in your machine
  • run the script demo.m and have fun

Authors

  • Xinyu Wang, Jiangxi Normal University
  • Hanxi Li, Jiangxi Normal University
  • Yi Li, Toyota Research Institute of North America
  • Fumin Shen, University of Electronic Science and Technology of China
  • Fatih Porikli, Australian National University, Data61

Citing MSDAT

If you find this repo useful in your research, please consider citing:

@article{wang2017robust,
  title={Robust and Real-time Deep Tracking Via Multi-Scale Domain Adaptation},
  author={Wang, Xinyu and Li, Hanxi and Li, Yi and Shen, Fumin and Porikli, Fatih},
  journal={arXiv preprint arXiv:1701.00561},
  year={2017}
}