Fashion Landmark Detection and Category Classification for Robotics

This repository contains the code and our in-lab dataset described in [1].

Authors Thomas Ziegler, Judith Butepage, Michael C. Welle, Anastasiia Varva, Tonci Novkovic and Danica Kragic

Maintainer Thomas Ziegler

The code builds up on the publicly available code from Liu and Lu [2].

Bibliography

The paper is accepted to ICARSC 2020 and will be published soon. If this code is used in a scientific publication please cite the following paper:

[1] T. Ziegler, J. Butepage, M. C. Welle, A Varva, T. Novokovic and D. Kagric, “Fashion Landmark Detection and Category Classification for Robotics”, in 2020 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC), 2020.

@incollection{ziegler2020fashion,
  title={Fashion Landmark Detection and Category Classification for Robotics},
  author={T. {Ziegler} and J. {Butepage} and M. C. {Welle} and A. {Varva} and T. {Novkovic} and D. {Kagric}}
  booktitle={2020 IEEE International Conference on Autonomous Robot Systems and Competitions (ICARSC)},
  year={2020},
}

[2] J. Liu, L. Hong, "Deep Fashion Analysis with Feature Map Upsampling and Landmark-Driven Attention, in European Conference on Computer Vision 2018 Workshops, 2018

Dependencies

• For the required python packages see conda.yml file

• Installation guide for the IORN see original repo

License

The source code is released under the GNU General Public License.