Automatic Discovery of Discriminative Parts as a Quadratic Assignment Problem

Abstract

Part-based image classification consists in representing categories by small sets of discriminative parts upon which a representation of the images is built. This paper addresses the question of how to automatically learn such parts from a set of labeled training images. We propose to cast the training of parts as a quadratic assignment problem in which optimal correspondences between image regions and parts are automatically learned. The paper analyses different assignment strategies and thoroughly evaluates them on two public datasets: Willow actions and MIT 67 scenes.

Cite

Text

Sicre et al. "Automatic Discovery of Discriminative Parts as a Quadratic Assignment Problem." IEEE/CVF International Conference on Computer Vision Workshops, 2017. doi:10.1109/ICCVW.2017.129

Markdown

[Sicre et al. "Automatic Discovery of Discriminative Parts as a Quadratic Assignment Problem." IEEE/CVF International Conference on Computer Vision Workshops, 2017.](https://mlanthology.org/iccvw/2017/sicre2017iccvw-automatic/) doi:10.1109/ICCVW.2017.129

BibTeX

@inproceedings{sicre2017iccvw-automatic,
  title     = {{Automatic Discovery of Discriminative Parts as a Quadratic Assignment Problem}},
  author    = {Sicre, Ronan and Rabin, Julien and Avrithis, Yannis and Furon, Teddy and Jurie, Frédéric and Kijak, Ewa},
  booktitle = {IEEE/CVF International Conference on Computer Vision Workshops},
  year      = {2017},
  pages     = {1059-1068},
  doi       = {10.1109/ICCVW.2017.129},
  url       = {https://mlanthology.org/iccvw/2017/sicre2017iccvw-automatic/}
}