Evaluation of Deep Image Descriptors for Texture Retrieval

Bojana Gajic, Eduard Vazquez, Ramon Baldrich

2017

Abstract

The increasing complexity learnt in the layers of a Convolutional Neural Network has proven to be of great help for the task of classification. The topic has received great attention in recently published literature. Nonetheless, just a handful of works study low-level representations, commonly associated with lower layers. In this paper, we explore recent findings which conclude, counterintuitively, the last layer of the VGG convolutional network is the best to describe a low-level property such as texture. To shed some light on this issue, we are proposing a psychophysical experiment to evaluate the adequacy of different layers of the VGG network for texture retrieval. Results obtained suggest that, whereas the last convolutional layer is a good choice for a specific task of classification, it might not be the best choice as a texture descriptor, showing a very poor performance on texture retrieval. Intermediate layers show the best performance, showing a good combination of basic filters, as in the primary visual cortex, and also a degree of higher level information to describe more complex textures.

Paper Citation

in Harvard Style

Gajic B., Vazquez E. and Baldrich R. (2017). Evaluation of Deep Image Descriptors for Texture Retrieval . In Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP, (VISIGRAPP 2017) ISBN 978-989-758-226-4, pages 251-257. DOI: 10.5220/0006129302510257

in Bibtex Style

@conference{visapp17,
author={Bojana Gajic and Eduard Vazquez and Ramon Baldrich},
title={Evaluation of Deep Image Descriptors for Texture Retrieval},
booktitle={Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP, (VISIGRAPP 2017)},
year={2017},
pages={251-257},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006129302510257},
isbn={978-989-758-226-4},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 12th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 5: VISAPP, (VISIGRAPP 2017)
TI - Evaluation of Deep Image Descriptors for Texture Retrieval
SN - 978-989-758-226-4
AU - Gajic B.
AU - Vazquez E.
AU - Baldrich R.
PY - 2017
SP - 251
EP - 257
DO - 10.5220/0006129302510257